2024-03-29T08:10:23Zhttps://iris.unitn.it/oai/requestoai:iris.unitn.it:11572/3773722024-03-10T21:30:44Zcom_11572_101873com_11572_101871col_11572_101864
Egocentric Hierarchical Visual Semantics
Luca ERCULIANI
Andrea BONTEMPELLI
Andrea PASSERINI
Fausto GIUNCHIGLIA
Erculiani, Luca
Bontempelli, Andrea
Passerini, Andrea
Giunchiglia, Fausto
Genus and Differentia, visual semantics, interactive machine learning
We are interested in aligning how people think about objects and what machines perceive, meaning by this the fact that object recognition, as performed by a machine, should follow a process which resembles that followed by humans when thinking of an object associated with a certain concept. The ultimate goal is to build systems which can meaningfully interact with their users, describing what they perceive in the users' own terms. As from the field of Lexical Semantics, humans organize the meaning of words in hierarchies where the meaning of, e.g., a noun, is defined in terms of the meaning of a more general noun, its genus, and of one or more differentiating properties, its differentia. The main tenet of this paper is that object recognition should implement a hierarchical process which follows the hierarchical semantic structure used to define the meaning of words. We achieve this goal by implementing an algorithm which, for any object, recursively recognizes its visual genus and its visual differentia. In other words, the recognition of an object is decomposed in a sequence of steps where the locally relevant visual features are recognized. This paper presents the algorithm and a first evaluation.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/377372
10.3233/FAIA230095
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85171436474
info:eu-repo/semantics/altIdentifier/isbn/978-1-64368-394-2
info:eu-repo/semantics/altIdentifier/isbn/978-1-64368-395-9
https://ebooks.iospress.nl/doi/10.3233/FAIA230095
eng
ispartofbook:Proceedings of the Second International Conference on Hybrid Human-Machine Intelligence
HHAI 2023
volume:368
firstpage:320
lastpage:329
numberofpages:10
serie:FRONTIERS IN ARTIFICIAL INTELLIGENCE AND APPLICATIONS
info:eu-repo/semantics/openAccess
IOS Press
place:Amsterdam
oai:iris.unitn.it:11572/3604852024-03-23T01:00:32Zcom_11572_101873com_11572_101871col_11572_101864
Capacity-Oriented RIS Control Through a Genetic Algorithm with Learning Capability
Zardi F.
Oliveri G.
Rocca P.
Massa A.
Zardi, F.
Oliveri, G.
Rocca, P.
Massa, A.
This work addresses the issue of controlling a Reconfigurable Intelligent Surface (RIS) in real-time to maximize the network capacity. A RIS control strategy is proposed exploiting a Genetic Algorithm (GA) augmented with learning capabilities to achieve high computation efficiency and guarantee consistent network performance in highly-dynamic scenarios. The proposed strategy does not require knowledge of the channel between the RIS and the users, enabling the design of RIS without channel-sensing hardware. The proposed strategy is demonstrated in a small-scale numerical example.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/360485
10.1109/AP-S/USNC-URSI47032.2022.9886649
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85139776907
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-9658-2
https://ieeexplore.ieee.org/document/9886649
eng
ispartofbook:2022 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting Proceedings
2022 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting (AP-S/URSI)
firstpage:69
lastpage:70
numberofpages:2
info:eu-repo/semantics/closedAccess
STAMPA
Institute of Electrical and Electronics Engineers
place:Piscataway, NJ
oai:iris.unitn.it:11572/3980942024-03-14T18:02:52Zcom_11572_101873com_11572_101871col_11572_101864
A deep multiple instance learning approach based on coarse labels for high-resolution land-cover mapping
Perantoni, Gianmarco
Bruzzone, Lorenzo
Bruzzone, Lorenzo
Bovolo, Francesca
Perantoni, Gianmarco
Bruzzone, Lorenzo
land-cover mapping, weak supervision, multiple instance learning, deep learning, remote sensing
The quantity and the quality of the training labels are central problems in high-resolution land-cover mapping
with machine-learning-based solutions. In this context, weak labels can be gathered in large quantities by
leveraging on existing low-resolution or obsolete products. In this paper, we address the problem of training
land-cover classifiers using high-resolution imagery (e.g., Sentinel-2) and weak low-resolution reference data (e.g.,
MODIS-derived land-cover maps). Inspired by recent works in Deep Multiple Instance Learning (DMIL), we
propose a method that trains pixel-level multi-class classifiers and predicts low-resolution labels (i.e., patch-level
classification), where the actual high-resolution labels are learned implicitly without direct supervision. This
is achieved with flexible pooling layers that are able to link the semantics of the pixels in the high-resolution
imagery to the low-resolution reference labels. Then, the Multiple Instance Learning (MIL) problem is re-framed
in a multi-class and in a multi-label setting. In the former, the low-resolution annotation represents the majority
of the pixels in the patch. In the latter, the annotation only provides us information on the presence of one of
the land-cover classes in the patch and thus multiple labels can be considered valid for a patch at a time, whereas
the low-resolution labels provide us only one label. Therefore, the classifier is trained with a Positive-Unlabeled
Learning (PUL) strategy. Experimental results on the 2020 IEEE GRSS Data Fusion Contest dataset show the
effectiveness of the proposed framework compared to standard training strategies.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/398094
10.1117/12.2679464
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85179554705
info:eu-repo/semantics/altIdentifier/isbn/9781510666955
info:eu-repo/semantics/altIdentifier/isbn/9781510666962
https://www.spiedigitallibrary.org/conference-proceedings-of-spie/12733/2679464/A-deep-multiple-instance-learning-approach-based-on-coarse-labels/10.1117/12.2679464.short?SSO=1
info:eu-repo/semantics/altIdentifier/wos/WOS:001118768500013
eng
ispartofbook:Image and Signal Processing for Remote Sensing XXIX
Image and Signal Processing for Remote Sensing XXIX, 2023
volume:12733
numberofpages:15
serie:PROCEEDINGS OF SPIE
info:eu-repo/semantics/openAccess
alleditors:Bruzzone, Lorenzo; Bovolo, Francesca
SPIE
place:Bellingham, Washington USA
oai:iris.unitn.it:11572/3857892024-03-23T00:45:53Zcom_11572_101873com_11572_101871col_11572_101864
Adaptive Expected Reactive algorithm for Heterogeneous Patrolling Systems based on Target Uncertainty
De Bona, Niccolò
Santoro, Luca
Brunelli, Davide
Fontanelli, Daniele
De Bona, Niccolò
Santoro, Luca
Brunelli, Davide
Fontanelli, Daniele
Multi-robot patrolling for dynamic coverage in flat environments is proposed, through a systematic simulative analysis between the Greedy Bayesian Strategy and the Expected Reactive algorithm based on the expected idleness. The two approaches are compared against unreliable communications, communication and sensing range, and number of conflicts. In addition, we introduce a new weighting-term for the regions close to a quantity of interest detected by robots, decreasing the passing-time for those regions. Combining the proposed control strategy and a traditional distributed and recursive Weighted Least Square estimation algorithm, the swarm is capable to compute the quantity of interest position with a desired target uncertainty. Extensive simulations and comparisons are reported.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/385789
10.1109/COMPSAC57700.2023.00017
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85168867045
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-2697-0
https://ieeexplore.ieee.org/document/10197120
info:eu-repo/semantics/altIdentifier/wos/WOS:001046484100007
eng
ispartofbook:2023 IEEE 47th Annual Computers, Software, and Applications Conference (COMPSAC)
COMPSAC 2023
firstpage:51
lastpage:56
numberofpages:6
info:eu-repo/semantics/closedAccess
IEEE
country:USA
place:Piscataway, New Jersey, USA
oai:iris.unitn.it:11572/3608822023-04-03T07:55:39Zcom_11572_101873com_11572_101871col_11572_101864
A Sound Up-to- n, δ Bisimilarity for PCTL
Bartoletti M.
Murgia M.
Zunino R.
Bartoletti, M.
Murgia, M.
Zunino, R.
We tackle the problem of establishing the soundness of approximate
bisimilarity with respect to PCTL and its relaxed semantics.
To this purpose, we consider a notion of bisimilarity similar to the one
introduced by Desharnais, Laviolette, and Tracol, which is parametric
with respect to an approximation error δ, and to the depth n of the
observation along traces. Essentially, our soundness theorem establishes
that, when a state q satisfies a given formula up-to error δ and steps n,
and q is bisimilar to q' up-to error δ' and enough steps, we prove that
q' also satisfies the formula up-to a suitable error δ'' and steps n. The
new error δ'' is computed from δ, δ' and the formula, and only depends
linearly on n. We provide a detailed overview of our soundness proof.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/360882
10.1007/978-3-031-08143-9_3
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85133012199
info:eu-repo/semantics/altIdentifier/isbn/978-3-031-08145-3
info:eu-repo/semantics/altIdentifier/isbn/978-3-031-08143-9
https://link.springer.com/chapter/10.1007/978-3-031-08143-9_3
info:eu-repo/semantics/altIdentifier/wos/WOS:000870760500003
eng
ispartofbook:Coordination Models and Languages
COORDINATION
volume:13271 LNCS
firstpage:35
lastpage:52
numberofpages:18
info:eu-repo/semantics/restrictedAccess
ELETTRONICO
Springer Cham
place:Springer Nature Switzerland
oai:iris.unitn.it:11572/3748312024-02-10T04:57:00Zcom_11572_101873com_11572_101871col_11572_101864
Movements by Drawing: Avant-Garde Set-Design and Sound Landscapes
Vattano, Starlight
C. L. Marcos
Vattano, Starlight
Drawing, dance, perception, set-design, Avant-garde theatre, futurist dance
The article deals with the study of visual images produced by the movement through the drawing’s set-design realized between 1920 and 1930, moment of theatre revolution, which saw the cooperation of different sectors of art, including photography, painting, architecture, cinema, dance, sculpture, and music. The visual values that triggered the theorization of the artistic avant-garde and, subsequently, the production of the scenographic drawing, in form and space, constitute a corollary of dynamic systems, a metaphor of the deconstructed movement through the figuration of the idea and at the same time signification of the interpretative distortion. The aim is to deepen the issue of movement expressed through the possible variations of the drawing. The study of the avant-garde set-designs that formed part of the structure characterizing the mobile theaters, allows to shed light on the role the graphic expression played in functional, compositional and aesthetic terms by self-propelled components, machines and images within the formalization process of the new theater. These elements constituted a new visual grammar no longer directed to the search for representative or symbolic values, but to the pure signification of itself, latent of a conceptualization process that traced a sign correspondence between object and subject in the drawing.
2019
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/374831
10.1007/978-3-319-93749-6_113
info:eu-repo/semantics/altIdentifier/isbn/978-3-319-93748-9
https://link.springer.com/chapter/10.1007/978-3-319-93749-6_113
eng
ispartofbook:Graphic Imprints. The Influence of Representation and Ideation Tools in Architecture
EGA 2018. XVII Congreso Internacional de Expresión Gráfica Arquitectónica
firstpage:1360
lastpage:1371
numberofpages:12
info:eu-repo/semantics/closedAccess
ELETTRONICO
alleditors:C. L. Marcos
Springer
country:CHE
place:Cham
oai:iris.unitn.it:11572/3535212023-05-26T12:44:40Zcom_11572_101873com_11572_101871col_11572_101864
Generative adversarial networks for subdomain enumeration
Degani L.
Bergadano F.
Mirheidari S. A.
Martinelli F.
Crispo B.
Degani, L.
Bergadano, F.
Mirheidari, S. A.
Martinelli, F.
Crispo, B.
Subdomain enumeration is a fundamental step of many security processes (i.e., vulnerability discovery, OSINT, host enumeration, etc.).
Up to now, this has been achieved with deterministic procedures that have shown some limitations. For instance, the process typically requires the generation of a candidate, which is subsequently
checked for validity. While the validation is a straightforward procedure, the definition of an optimal candidate generation strategy is still an open problem. This paper presents a novel subdomain
enumeration tool that allows the generation of high-quality subdomain candidates. We employ a Generative Adversarial Network (GAN) to sample unseen candidates from the distribution of valid
subdomain names. The model learns this distribution from publicly available datasets. Moreover, by sampling from the trained model, we address the limitations of traditional algorithms. Our
experiments were carried out against 15 domains and a ground truth of 1164 other targets. The 15 domains were carefully selected from bug bounty platforms to avoid terms of use violations. Several factors influenced the choices, including the popularity, the expected number of subdomains, and the available services. Our experiments aim to validate our approach by testing the performance increase in subdomain enumeration processes against the state-of-the-art. We benchmark our proposal in terms of candidates’ validity and sample uniqueness. The results showed that, with our GAN, the performance of a traditional subdomain enumeration workflow increased by up to 61%. In addition, according to our
ground truth experiments, the GAN was able to guess, on average, 32% of subdomains.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/353521
10.1145/3477314.3506967
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85130417044
info:eu-repo/semantics/altIdentifier/isbn/9781450387132
https://dl.acm.org/doi/10.1145/3477314.3506967
info:eu-repo/semantics/altIdentifier/wos/WOS:000946564100223
eng
ispartofbook:Proceedings of the 37th ACM/SIGAPP Symposium on Applied Computing
The 37th ACM/SIGAPP Symposium On Applied Computing
firstpage:1636
lastpage:1645
numberofpages:10
info:eu-repo/semantics/openAccess
Association for Computing Machinery
place:New York, USA
oai:iris.unitn.it:11572/3748672024-02-10T04:56:48Zcom_11572_101873com_11572_101871col_11572_101864
Configuración smart para el patrimonio Euro-Mediterráneo: el dibujo smart para el Rabàto de Agrigento
Vattano, Starlight
Garcia Bujalance
Susana
Vargas Yáñez, Antonio, D
Vattano, Starlight
Garcia, Bujalance
Susana, Alberto
Rehabilitation, suburbs, historic city, Agrigento (Rehabilitación, suburbs, ciudad histórica, Agrigento)
Research on the rehabilitation of a district of the suburbs of the historic Italian city of Agrigento, from reinterpreting historical frame in a new key to space utilization invites reflection on the concept of urban regeneration in historic areas medium cities of Euro-Mediterranean context, at a time when the growth of cities seems to enter a phase of stagnation in which the tourism component of these areas reveals a possibility of urban transformation. The article proposes a methodology for analysis and intervention Rabato district in which energy efficiency, implementation of smart technologies and citizen participation are complemented by the incorporation of economic, cultural and social values, to build an expanded concept of smart city. In addition, we propose the creation of a mapping system and visualization of urban data, which may constitute a network of historic smart cities, which serve to transfer and share knowledge and intervention methodologies.
La investigación sobre la rehabilitación de un barrio de los arrabales de la ciudad histórica italiana de Agrigento a partir de la reinterpretación de su trama histórica en una nueva clave de aprovechamiento del espacio, invita a reflexionar sobre el concepto de regeneración urbana en áreas históricas de ciudades medias del contexto euro mediterráneo, en un momento en el que el crecimiento de las ciudades parece entrar en una fase de estancamiento y en el que la componente turística de estos ámbitos se revela como una posibilidad de transformación urbana. El artículo propone una metodología sobre el análisis y la intervención en el barrio del Rabato en el que la eficiencia energética, la implementación de tecnologías smart y la participación ciudadana, se complementan con la incorporación de los valores patrimoniales, culturales y sociales para construir un concepto expandido de ciudad inteligente. Además, se propone la creación de un sistema de mapeo y visibilización de datos urbanos, que pudieran constituir una red de ciudades históricas smart, que sirviese para transferir y compartir conocimiento y metodologías de intervención.
2014
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/374867
info:eu-repo/semantics/altIdentifier/isbn/978-84-697-0799-9
eng
spa
ispartofbook:Convocatoria de Comunicaciones y 2a Bienal de Proyectos de Edificación y Urbanismo Sostenible
Greencities y Sostenibilidad. Convocatoria de Comunicaciones y 2a Bienal de Proyectos de Edificación y Urbanismo Sostenible.
firstpage:381
lastpage:404
numberofpages:24
info:eu-repo/semantics/closedAccess
STAMPA
alleditors:Vargas Yáñez, Antonio, D
Greencities y Sostenibilidad
country:ESP
place:Malaga
oai:iris.unitn.it:11572/3906292024-01-23T00:15:48Zcom_11572_101873com_11572_101871col_11572_101864
Soundscape – Paesaggi sonori nelle aule di scuola primaria
Visentin, Chiara
Torresin, Simone
Pellegatti, Matteo
Prodi, Nicola
Visentin, Chiara
Torresin, Simone
Pellegatti, Matteo
Prodi, Nicola
Il soundscape degli ambienti interni è un tema di ricerca emergente, strettamente connesso alla progettazione di spazi di vita che supportano il benessere degli occupanti. Questo studio pilota si focalizza sulle aule della scuola primaria, in cui la percezione dell’ambiente sonoro da parte dei bambini (n = 130, 8-10 anni) è stata indagata utilizzando scale grafiche, in riferimento a dominanza e frequenza degli stimoli sonori e alle dimensioni affettive di piacevolezza e attivazione.
SOMMARIO
Il soundscape degli ambienti interni è un tema di ricerca emergente, strettamente connesso alla progettazione di spazi di vita che supportano il benessere degli occupanti. Questo studio pilota si focalizza sulle aule della scuola primaria, in cui la percezione dell’ambiente sonoro da parte dei bambini (n = 130, 8-10 anni) è stata indagata utilizzando scale grafiche, in riferimento a dominanza e frequenza degli stimoli sonori e alle dimensioni affettive di piacevolezza e attivazione.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/390629
info:eu-repo/semantics/altIdentifier/isbn/978-88-88942-66-7
ita
ispartofbook:Associazione Italiana di Acustica 49° Convegno Nazionale
Associazione Italiana di Acustica 49° Convegno Nazionale Ferrara
numberofpages:2
info:eu-repo/semantics/openAccess
Associazione Italiana di Acustica
place:Ferrara
oai:iris.unitn.it:11572/4016682024-03-20T01:38:56Zcom_11572_101873com_11572_101871col_11572_101864
Progetto TRE3: sistema ibrido a pareti intelaiate in legno per l’edilizia di emergenza
Lorenzo Franzoni
Matteo Izzi
Albino Angeli
Ester Sinitoa
Daniele Casagrande
Gaia Pasetto
Andrea Polastri
Franzoni, Lorenzo
Izzi, Matteo
Angeli, Albino
Sinitoa, Ester
Casagrande, Daniele
Pasetto, Gaia
Polastri, Andrea
Strutture in legno, edilizia di emergenza, pareti intelaiate, CLT, caratterizzazione meccanica
Il Progetto TRE3 si focalizza sullo sviluppo di un sistema strutturale ibrido a pareti intelaiate in legno (nel seguito
indicato come Hybrid Timber-Frame, HTF). L’HTF accoppia i vantaggi di due dei sistemi costruttivi in legno più
diffusi sia in Italia che in Europa: la leggerezza e la capacità dissipativa del telaio leggero (Light Timber-Frame,
LTF) e le prestazione meccaniche dei pannelli di tavole incrociate (Cross-Laminated Timber, CLT, in Italia anche
conosciuto come X-Lam). In condizioni di emergenza, l’impiego di moduli HTF permette un rapido assemblaggio
di unità abitative prefabbricate, nonché ammette la possibilità di smontaggio e successivo rimontaggio in altra sede.
TRE3 si propone come principale obiettivo l’analisi meccanica e strutturale del sistema HTF; parallelamente a ciò,
sono tenuti in considerazione gli aspetti multidisciplinari legati alla logistica, al confort abitativo e ai dettagli architettonici. Il presente articolo affronta la definizione dei moduli parete usati per la realizzazione di unità abitative di
emergenza e discute i risultati ottenuti da test di caratterizzazione meccanica volti ad indagarne il comportamento
in presenza di azioni sismiche.
2017
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/401668
ita
ispartofbook:ANIDIS 2017
ANIDIS 2017
numberofpages:8
info:eu-repo/semantics/closedAccess
ANIDIS 2017
place:Pistoia (Italia)
oai:iris.unitn.it:11572/4009982024-03-23T01:04:21Zcom_11572_101873com_11572_101871col_11572_101864
Tightening Classification Boundaries in Open Set Domain Adaptation through Unknown Exploitation
e Silva, Lucas Fernando Alvarenga
Sebe, Nicu
Almeida, Jurandy
e Silva, Lucas Fernando Alvarenga
Sebe, Nicu
Almeida, Jurandy
Training, Graphics, Adaptation models,
Benchmark testing, Data augmentation,
Convolutional neural networks,
Task analysis
Convolutional Neural Networks (CNNs) have brought revolutionary advances to many research areas due to their capacity of learning from raw data. However, when those methods are applied to non-controllable environments, many different factors can degrade the model's expected performance, such as unlabeled datasets with different levels of domain shift and category shift. Particularly, when both issues occur at the same time, we tackle this challenging setup as Open Set Domain Adaptation (OSDA) problem. In general, existing OSDA approaches focus their efforts only on aligning known classes or, if they already extract possible negative instances, use them as a new category learned with supervision during the course of training. We propose a novel way to improve OSDA approaches by extracting a high-confidence set of unknown instances and using it as a hard constraint to tighten the classification boundaries of OSDA methods. Especially, we adopt a new loss constraint evaluated in three different means, (1) directly with the pristine negative instances; (2) with randomly transformed negatives using data augmentation techniques; and (3) with synthetically generated negatives containing adversarial features. We assessed all approaches in an extensive set of experiments based on OVANet, where we could observe consistent improvements for two public benchmarks, the Office-31 and Office-Home datasets, yielding absolute gains of up to 1.3% for both Accuracy and H-Score on Office-31 and 5.8% for Accuracy and 4.7% for H-Score on Office-Home.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/400998
10.1109/SIBGRAPI59091.2023.10347139
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-3872-0
https://ieeexplore.ieee.org/document/10347139/authors#authors
eng
ispartofbook:36th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)
SIBGRAPI
firstpage:157
lastpage:162
numberofpages:6
serie:SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)
info:eu-repo/semantics/openAccess
IEEE
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3700312023-04-18T15:39:43Zcom_11572_101873com_11572_101871col_11572_101864
Using Pre-Trained Language Models for Producing Counter Narratives Against Hate Speech: a Comparative Study
Tekiroglu Serra Sinem
Bonaldi Helena
Fanton Margherita
Guerini Marco
Smaranda, Muresan
Preslav, Nakov
Aline, Villavicencio
Tekiroglu, Serra Sinem
Bonaldi, Helena
Fanton, Margherita
Guerini, Marco
In this work, we present an extensive study on the use of pre-trained language models for the task of automatic Counter Narrative (CN) generation to fight online hate speech in English. We first present a comparative study to determine whether there is a particular Language Model (or class of LMs) and a particular decoding mechanism that are the most appropriate to generate CNs. Findings show that autoregressive models combined with stochastic decodings are the most promising. We then investigate how an LM performs in generating a CN with regard to an unseen target of hate. We find out that a key element for successful {`}out of target{'} experiments is not an overall similarity with the training data but the presence of a specific subset of training data, i. e. a target that shares some commonalities with the test target that can be defined a-priori. We finally introduce the idea of a pipeline based on the addition of an automatic post-editing step to refine generated CNs.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/370031
10.18653/v1/2022.findings-acl.245
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85140792267
info:eu-repo/semantics/altIdentifier/isbn/978-1-955917-25-4
https://aclanthology.org/2022.findings-acl.245
info:eu-repo/semantics/altIdentifier/wos/WOS:000828767403016
eng
ispartofbook:Findings of the Association for Computational Linguistics: ACL 2022
Annual Meeting of the Association for Computational Linguistics (ACL)
firstpage:3099
lastpage:3114
numberofpages:16
info:eu-repo/semantics/openAccess
alleditors:Smaranda, Muresan; Preslav, Nakov; Aline, Villavicencio
Association for Computational Linguistics
place:209 N. Eighth Street, Stroudsburg PA 18360, USA
oai:iris.unitn.it:11572/3533792024-01-17T00:09:53Zcom_11572_101873com_11572_101871col_11572_101864
Model-based parameter analysis of dielectric elastomer loudspeakers
Moretti, Giacomo
Rizzello, Gianluca
Moretti, Giacomo
Rizzello, Gianluca
Dielectric elastomers (DEs) are polymeric multi-functional materials that can be used to develop lightweight electrostatic actuators. Among other, DEs allow developing coilfree loudspeakers, in which the acoustic diaphragm and the actuator are embedded into a single DE membrane, whose deformations are driven by electrostatic stresses.In this paper, we present a simulation analysis of a DE loudspeaker with the aim of highlighting the effect of relevant design parameters (namely, the DE membrane thickness, diameter, and geometric aspect ratio) on the acoustic response. For the sake of illustration, we make reference to a simple loudspeaker layout, which does not require any mechanical or pneumatic biasing elements, and only consists in a DE membrane pre-loaded off-plane. Based on a validated finite element multi-physics model of the system, we discuss how the system response (namely, eigenfrequencies and generated sound pressure level) varies with the design parameters. The presented results point out thresholds and trends that are relevant for the choice of DE loudspeakers’ design parameters.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/353379
10.1109/AIM52237.2022.9863404
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85137678084
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-1308-4
info:eu-repo/semantics/altIdentifier/isbn/9781665413091
https://ieeexplore.ieee.org/document/9863404
info:eu-repo/semantics/altIdentifier/wos/WOS:000860750800116
eng
ispartofbook:2022 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM)
2022 IEEE/ASME AIM
firstpage:1129
lastpage:1135
numberofpages:7
serie:IEEE/ASME INTERNATIONAL CONFERENCE ON ADVANCED INTELLIGENT MECHATRONICS
info:eu-repo/semantics/openAccess
IEEE/ASME
country:USA
place:Piscataway, NJ
oai:iris.unitn.it:11572/4015712024-03-16T02:15:57Zcom_11572_101873com_11572_101871col_11572_101864
Lateral Deformation and Kinematic Modes of Multipanel Balloon-Type Clt Shearwall System
Xing D.
Doudak G.
Casagrande D.
Nyrud, A. Q.
Malo, K. A. ... [et al.]
Xing, D.
Doudak, G.
Casagrande, D.
Aspect ratio
Balloon-type
Cross-laminated timber
Kinematic behaviour
Multi-panel
Balloon-type Cross-Laminated Timber (CLT) shearwall systems have been widely used as lateral load resisting systems (LLRS) in several mid- and high-rise timber buildings. Contrary to platform-type shearwalls, balloon framing contains continuous panels along multiple storeys in the vertical direction, which helps avoid perpendicular-to-grain compression failure in the floors. Despite some of the advantages associated with this type of construction, design provisions have not yet been introduced to various timber codes and standards, mainly due to the scarcity in research on their behaviours. This paper presents a numerical analysis model with the aim of investigating the lateral deformation and kinematic modes of balloon-type CLT shearwall systems. A finite element (FE) model is proposed, and a discussion on various key parameters affecting the behaviour of the shearwall, including aspect ratios (h/b), stiffness of the connectors, vertical loads and the number of panels, is presented.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/401571
10.52202/069179-0314
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85171452826
info:eu-repo/semantics/altIdentifier/isbn/978-1-7138-7329-7
info:eu-repo/semantics/altIdentifier/isbn/978-1-7138-7327-3
eng
ispartofbook:13th World Conference on Timber Engineering, WCTE 2023
WCTE 2023
volume:4
firstpage:2376
lastpage:2383
numberofpages:8
info:eu-repo/semantics/closedAccess
alleditors:Nyrud, A. Q.; Malo, K. A. ... [et al.]
World Conference on Timber Engineering (WCTE)
place:Oslo
oai:iris.unitn.it:11572/3612682023-04-03T07:43:32Zcom_11572_101873com_11572_101871col_11572_101864
Predict, Prevent, and Evaluate: Disentangled Text-Driven Image Manipulation Empowered by Pre-Trained Vision-Language Model
Xu Z.
Lin T.
Tang H.
Li F.
He D.
Sebe N.
Timofte R.
Van Gool L.
Ding E.
Xu, Z.
Lin, T.
Tang, H.
Li, F.
He, D.
Sebe, N.
Timofte, R.
Van Gool, L.
Ding, E.
Face and gesture
Image and video synthesis and generation
To achieve disentangled image manipulation, previous works depend heavily on manual annotation. Meanwhile, the available manipulations are limited to a pre-defined set the models were trainedfor. We propose a novelframework, i.e., Predict, Prevent, and Evaluate (PPE), for disentangled text-driven image manipulation that requires little manual annotation while being applicable to a wide variety of ma-nipulations. Our method approaches the targets by deeply exploiting the power of the large-scale pre-trained vision-language model CLIP [32]. Concretely, we firstly Predict the possibly entangled attributes for a given text command. Then, based on the predicted attributes, we introduce an entanglement loss to Prevent entanglements during training. Finally, we propose a new evaluation metric to Evaluate the disentangled image manipulation. We verify the effectiveness of our method on the challenging face editing task. Extensive experiments show that the proposed PPE frame-work achieves much better quantitative and qualitative re-sults than the up-to-date StyleCLIP [31] baseline. Code is available at https://github.com/zipengxuc/PPE.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/361268
10.1109/CVPR52688.2022.01769
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85137433390
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-6946-3
https://ieeexplore.ieee.org/document/9878500/authors#authors
info:eu-repo/semantics/altIdentifier/wos/WOS:000870783004004
eng
ispartofbook:Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
volume:2022
firstpage:18208
lastpage:18217
numberofpages:10
serie:PROCEEDINGS IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION
info:eu-repo/grantAgreement/EC/H2020/951911
info:eu-repo/semantics/openAccess
IEEE Computer Society
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3550212024-03-06T05:28:29Zcom_11572_101873com_11572_101871col_11572_101864
An empirical tool to classify industries by regional concentration and spatial polarization
Giuliani, Diego
Dickson, Maria Michela
Santi, Flavio
Espa, Giuseppe
Balzanella, A
Bini, M.
Cavicchia, C.
Verde, R.
Giuliani, Diego
Dickson, Maria Michela
Santi, Flavio
Espa, Giuseppe
Traditional measures of geographical concentration of industries based on regional data (such as Gini, Herfindhal, and Ellison-Glaeser indices) do not consider the information about the spatial positions of regions. This implies their insensitivity to regions’ spatial order and inability to account for neighboring effects. As an attempt to cope with this limitation, a recent streamof literature [7, 4, 9, among others] has focused on developing measures that quantify the degree of concentration of an industry while adjusting for spatial connections among regions. Following the idea that a single measure cannot fully describe the characteristics of an industry in terms of both concentration and spatial interactions, this paper proposes an alternative approach that measures the two dimensions jointly and allows for the classification of economic sectors into meaningful types of geographical configurations.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/355021
info:eu-repo/semantics/altIdentifier/isbn/9788891932310
https://meetings3.sis-statistica.org/index.php/itacosm2023/Caserta
eng
ita
ispartofbook:51st Scientific Meeting of the Italian Statistical Society: Book of the Short Papers
SIS 2022
firstpage:340
lastpage:345
numberofpages:6
info:eu-repo/semantics/closedAccess
alleditors:Balzanella, A; Bini, M.; Cavicchia, C.; Verde, R.
Pearson
place:Milano
oai:iris.unitn.it:11572/3520602024-03-10T11:23:03Zcom_11572_101873com_11572_101871col_11572_101864
MULTIOBJECTIVE ANALYSIS OF OPEN AREAS INVADED BY FOREST WITH OPEN-SOURCE SOFTWARE: THE CASE OF THE SATURN PROJECT
Pianegonda, A.
Frisanco, F.
Betta, A.
Ciolli, M.
Brovelli M. A., Ciolli M. , Minghini M.
Pianegonda, A.
Frisanco, F.
Betta, A.
Ciolli, M.
multi-objective analysis, multi-criteria analysis, FOSS4G software, agricultural areas, forests, open areas
In the last decades in Italian mountainous regions, forests are invading abandoned pastures and cultivated surfaces that often played a key ecological role for biodiversity conservation, a complex land-use change phenomenon. To improve sustainable regional planning and management it is increasingly important to quantify the phenomenon and classify those areas according their ecological vocation. This study explores multi-objective and multi-criteria assessment for the identification of the most suitable areas for agricultural purposes between those surfaces that have been invaded by forests. Free and Open-Source Software for Geospatial (FOSS4G) software has been used to carry out the research. The work here presented has been funded by the EIT Climate-KIC SATURN project (2018-2021) and was carried out in the pilot areas of the Province of Trento (Italy). Geospatial data set was georeferenced and managed with GRASS and QGIS and the files were collected combining data freely available at the Autonomous Province of Trento geocatalogue as well as others self-produced during the project. The comparative analysis and methodology were carried out by means of QGIS 3.16 Geographic Information System which has been used to complete the analysis to develop a methodology that can be widely used by territorial operators and Public Administrations. To validate the model and verify the results, on-site inspections were carried out. The developed model can be used in different environmental assessments for territorial planning purposes.
2022
info:eu-repo/semantics/conferenceObject
http://hdl.handle.net/11572/352060
10.5194/isprs-archives-XLVIII-4-W1-2022-357-2022
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85137179765
https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLVIII-4-W1-2022/357/2022/
eng
ispartofbook:Volume XLVIII-4/W1-2022, 2022 | ISPRS WG IV/4, III/10, IV/7, V/1 & ICWG IV/III Free and Open Source Software for Geospatial (FOSS4G) 2022 – Academic Track
FOSS4G 2022 Firenze
volume:XLVIII-4/W1-2022
firstpage:357
lastpage:363
numberofpages:7
journal:INTERNATIONAL ARCHIVES OF THE PHOTOGRAMMETRY, REMOTE SENSING AND SPATIAL INFORMATION SCIENCES
info:eu-repo/semantics/openAccess
ELETTRONICO
alleditors:Brovelli M. A., Ciolli M. , Minghini M.
Copernicus GmbH
country:DEU
place:Gottingen, Germany
oai:iris.unitn.it:11572/4007652024-03-15T17:43:39Zcom_11572_101873com_11572_101871col_11572_101864
KG-SP: Knowledge Guided Simple Primitives for Open World Compositional Zero-Shot Learning
Karthik S.
Mancini M.
Akata Z.
Karthik, S.
Mancini, M.
Akata, Z.
categorization
Recognition: detection
Representation learning
retrieval
Transfer/low-shot/long-tail learning
The goal of open-world compositional zero-shot learning (OW-CZSL) is to recognize compositions of state and objects in images, given only a subset of them during training and no prior on the unseen compositions. In this setting, models operate on a huge output space, containing all possible state-object compositions. While previous works tackle the problem by learning embeddings for the compositions jointly, here we revisit a simple CZSL baseline and predict the primitives, i.e. states and objects, independently. To ensure that the model develops primitive-specific features, we equip the state and object classifiers with separate, non-linear feature extractors. Moreover, we estimate the feasibility of each composition through external knowledge, using this prior to remove unfeasible compositions from the output space. Finally, we propose a new setting, i.e. CZSL under partial supervision (pCZSL), where either only objects or state labels are available during training, and we can use our prior to estimate the missing labels. Our model, Knowledge-Guided Simple Primitives (KG-SP), achieves state of the art in both OW-CZSL and pCZSL, surpassing most recent competitors even when coupled with semi-supervised learning techniques. Code available at: https://github.com/ExplainableML/KG-SP.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/400765
10.1109/CVPR52688.2022.00912
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85141800141
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-6946-3
https://ieeexplore.ieee.org/document/9878578/authors#authors
info:eu-repo/semantics/altIdentifier/wos/WOS:000870759102039
eng
ispartofbook:2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
firstpage:9326
lastpage:9335
numberofpages:10
serie:PROCEEDINGS IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION
info:eu-repo/semantics/openAccess
IEEE Computer Society
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3768742024-03-10T03:18:18Zcom_11572_101873com_11572_101871col_11572_101864
From a hydroelectric power centre to an arts centre: the regeneration of the Fies hydroelectric power station in the Sarca valley
Gatti, Maria Paola
Dalla Serra, Deanna
Cacciaguerra, Giorgio
Carmine Gambardella
Gatti, Maria Paola
Dalla Serra, Deanna
Cacciaguerra, Giorgio
regeneration, re-use, re cycle, abandoned industrial buildings, intervention strategy
Many production buildings have been abandoned; few have had a new life after being closed down.
Industrial buildings are abandoned for many reasons. These are sometimes related to the specific
sector: for example, economic and production transformation or a decline in the quantity of materials
etc., and sometimes they are specifically related to the building itself — to its technical life cycle — but
also related to general factors tied to changes in policy, for example, a different use of the water supply
or environmental factors.
For some disused industrial buildings, sustainable regeneration projects have been proposed and
implemented by interpreting and recovering the value of the memory of the work, the architecture and
the technology. We thought it would be interesting to analyse the renovation of the Fies hydroelectric
power station located on the right bank of the Sarca river (Trento), which today is a space for the creation
and production of contemporary arts.
2021
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/376874
info:eu-repo/semantics/altIdentifier/isbn/978-88-492-4088-7
eng
ispartofbook:WORLD HERITAGE and DESIGN FOR HEALTH
Le vie dei mercanti
firstpage:752
lastpage:758
numberofpages:7
STAMPA
alleditors:Carmine Gambardella
Gangemi Editore
country:ITA
place:Roma
oai:iris.unitn.it:11572/3978992024-02-01T15:08:21Zcom_11572_101873com_11572_101871col_11572_101864
Dynamically Instance-Guided Adaptation: A Backward-free Approach for Test-Time Domain Adaptive Semantic Segmentation
Wang, Wei
Zhong, Zhun
Wang, Weijie
Chen, Xi
Ling, Charles
Wang, Boyu
Sebe, Nicu
Wang, Wei
Zhong, Zhun
Wang, Weijie
Chen, Xi
Ling, Charle
Wang, Boyu
Sebe, Nicu
Adaptation models, Dams, Semantic segmentation, Source coding,
Semantics, Prototypes, Object detection
In this paper, we study the application of Test-time domain adaptation in semantic segmentation (TTDA-Seg) where both efficiency and effectiveness are crucial. Existing methods either have low efficiency (e.g., backward optimization) or ignore semantic adaptation (e.g., distribution alignment). Besides, they would suffer from the accumulated errors caused by unstable optimization and abnormal distributions. To solve these problems, we propose a novel backward-free approach for TTDA-Seg, called Dynamically Instance-Guided Adaptation (DIGA). Our principle is utilizing each instance to dynamically guide its own adaptation in a non-parametric way, which avoids the error accumulation issue and expensive optimizing cost. Specifically, DIGA is composed of a distribution adaptation module (DAM) and a semantic adaptation module (SAM), enabling us to jointly adapt the model in two indispensable aspects. DAM mixes the instance and source BN statistics to encourage the model to capture robust representation. SAM combines the historical prototypes with instance-level prototypes to adjust semantic predictions, which can be associated with the parametric classifier to mutually benefit the final results. Extensive experiments evaluated on five target domains demonstrate the effectiveness and efficiency of the proposed method. Our DIGA establishes new state-of-the-art performance in TTDA-Seg. Source code is available at: https://github.com/Waybaba/DIGA.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/397899
10.1109/CVPR52729.2023.02307
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-0129-8
https://ieeexplore.ieee.org/document/10204752/authors#authors
info:eu-repo/semantics/altIdentifier/wos/WOS:001062531308041
eng
ispartofbook:IEEE/CVF Conference on Computer Vision and Pattern Recognition
IEEE/CVF Conference on Computer Vision and Pattern Recognition
firstpage:24090
lastpage:24099
numberofpages:10
serie:PROCEEDINGS IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION
info:eu-repo/semantics/openAccess
IEEE
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/4040902024-03-06T11:37:16Zcom_11572_101873com_11572_101871col_11572_101864
The clothoid: a historical, literary and artistic introduction with applications to technology
Bertolazzi, Enrico
Frego, Cecilia
Frego, Marco
Hosseini, Seyed Mohsen
Peer, Angelika
Bertolazzi, Enrico
Frego, Cecilia
Frego, Marco
Hosseini, Seyed Mohsen
Peer, Angelika
In this work, we trace the history of the clothoid curve, a peculiar spiral with several technological applications. Starting from the bibliographic research by Levien, we deepen its history, focusing our attention on Cesaro's and Markov's studies of the curve. We also investigate the use of clothoids in modern times, highlighting its extension into three dimensions. Finally, we focus on the significance of the computation of the Fresnel integrals, which proved to be key for overcoming the historical bottleneck and paving the way for its practical use in applications.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/404090
10.1109/histelcon56357.2023.10365736
https://2023.ieee-histelcon.org
eng
ispartofbook:IEEE History of Telecommunications Conference, HISTELCON
HISTELCON
IEEE Xplore
place:USA
oai:iris.unitn.it:11572/3545652023-04-03T07:57:09Zcom_11572_101873com_11572_101871col_11572_101864
Gender Recognition from 3D Shape Parameters
Martinelli G.
Garau N.
Conci N.
Martinelli, G.
Garau, N.
Conci, N.
Body shape
Gender recognition
Parametric human body model
Gender recognition from images is generally approached by extracting the salient visual features of the observed subject, either focusing on the facial appearance or by analyzing the full body. In real-world scenarios, image-based gender recognition approaches tend to fail, providing unreliable results. Face-based methods are compromised by environmental conditions, occlusions (presence of glasses, masks, hair), and poor resolution. Using a full-body perspective leads to other downsides: clothing and hairstyle may not be discriminative enough for classification, and background cluttering could be problematic. We propose a novel approach for body-shape-based gender classification. Our contribution consists in introducing the so-called Skinned Multi-Person Linear model (SMPL) as 3D human mesh. The proposed solution is robust to poor image resolution and the number of features for the classification is limited, making the recognition task computationally affordable, especially in the classification stage, where less complex learning architectures can be easily trained. The obtained information is fed to an SVM classifier, trained and tested using three different datasets, namely (i) FVG, containing videos of walking subjects (ii) AMASS, collected by converting MOCAP data of people performing different activities into realistic 3D human meshes, and (iii) SURREAL, characterized by synthetic human body models. Additionally, we demonstrate that our approach leads to reliable results even when the parametric 3D mesh is extracted from a single image. Considering the lack of benchmarks in this area, we trained and tested the FVG dataset with a pre-trained Resnet50, for comparing our model-based method with an image-based approach.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/354565
10.1007/978-3-031-13324-4_18
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85136133604
info:eu-repo/semantics/altIdentifier/isbn/978-3-031-13323-7
info:eu-repo/semantics/altIdentifier/isbn/978-3-031-13324-4
https://link.springer.com/chapter/10.1007/978-3-031-13324-4_18#citeas
info:eu-repo/semantics/altIdentifier/wos/WOS:000870536000018
eng
ispartofbook:Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
21st International Conference on Image Analysis and Processing , ICIAP 2022
volume:13374
firstpage:203
lastpage:214
numberofpages:12
serie:LECTURE NOTES IN COMPUTER SCIENCE
info:eu-repo/semantics/embargoedAccess
ELETTRONICO
Springer Science and Business Media Deutschland GmbH
country:DEU
place:Cham, Svizzera
oai:iris.unitn.it:11572/4007292024-03-17T01:02:05Zcom_11572_101873com_11572_101871col_11572_101864
A Neuro-Symbolic Approach for Non-Intrusive Load Monitoring
Apriceno, G.
Erculiani, L.
Passerini, Andrea
Apriceno, G.
Erculiani, L.
Passerini, Andrea
A requirement of Smart Grids is the ability to predict the energy consumption patterns of their users. In the residential domain, this is usually not feasible due to the inability of the grid to dialog with (legacy) domestic appliances. To overcome this issue Non Intrusive Load Monitoring (NILM) was introduced, a task in which a predictor is used to disaggregate household power consumption. Many of the newer approaches make use of Neural Networks to accomplish this task, due to their superior ability to detect patterns in temporal (thus sequential) data. These models unfortunately require a huge amount of data to achieve good performance, and have the tendency to overfit the training data, making them difficult to predict future consumptions. For these reasons, adapting them to optimally predict a (future) house's consumption requires expensive and often prohibitive data collection phases. We propose a solution in the form of a neuro-symbolic framework that refines neural network predictions via a constrained optimization problem modelling the characteristics of the appliances of a house. This combined approach achieves superior performance with respect to the neural network alone over two out of five appliances and comparable results for the remaining ones, without requiring further training data.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/400729
10.3233/FAIA230638
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85175802717
info:eu-repo/semantics/altIdentifier/isbn/9781643684369
info:eu-repo/semantics/altIdentifier/isbn/9781643684376
https://ebooks.iospress.nl/doi/10.3233/FAIA230638
eng
ispartofbook:26th European Conference on Artificial Intelligence
26th European Conference on Artificial Intelligence, ECAI 2023
volume:372
firstpage:3175
lastpage:3181
numberofpages:7
serie:FRONTIERS IN ARTIFICIAL INTELLIGENCE AND APPLICATIONS
info:eu-repo/semantics/openAccess
IOS Press BV
place:online
oai:iris.unitn.it:11572/4036702024-03-21T19:15:48Zcom_11572_101873com_11572_101871col_11572_101864
Investigation into the Breakdown of Continuum Fracture Mechanics at the Nanoscale: Synthesis of Recent Results on Silicon
Gallo P
Sumigawa T
Shimada T
Kitamura T
Correia, José
de Jesus,Abílio M.P.
Gallo, P
Sumigawa, T
Shimada, T
Kitamura, T
The present contribution reviews some recent results on the experimental characterisation of the nanoscale fracture toughness of silicon by using pre-cracked specimens and alternatively the theory of critical distances (TCD). Later, the results are discussed to provide the ultimate dimensional limit of the continuum fracture mechanics at the nanoscale in the light of sophisticated discrete atomic simulations at the onset of brittle fracture. The results show that the fracture toughness of Si is independent of the scale, crystal orientation and the singular stress field length. This confirms the atomistic nature of the brittle fracture. Moreover, the continuum fracture mechanics fails below a singular stress field approaching 2 nm.
2018
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/403670
10.1007/978-3-319-91989-8_45
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85068512737
info:eu-repo/semantics/altIdentifier/isbn/978-3-319-91988-1
info:eu-repo/semantics/altIdentifier/isbn/978-3-319-91989-8
https://link.springer.com/chapter/10.1007/978-3-319-91989-8_45#citeas
info:eu-repo/semantics/altIdentifier/wos/WOS:000521737400045
eng
ispartofbook:ICTAEM 2018: Proceedings of the First International Conference on Theoretical, Applied and Experimental Mechanics
ICTAEM
volume:5
firstpage:205
lastpage:210
numberofpages:6
serie:STRUCTURAL INTEGRITY
ELETTRONICO
alleditors:Correia, José; de Jesus,Abílio M.P.;
Springer, Cham
place:Switzerland
oai:iris.unitn.it:11572/3733272024-03-23T00:58:31Zcom_11572_101873com_11572_101871col_11572_101864
Hemp Concrete Walls: Evaluation of the Relationship between CO2 and TVOC
Lara-Ibeas, Irene
Torresin, Simone
Ricciuti, Silvia
Babich, Francesco
Lara-Ibeas, Irene
Torresin, Simone
Ricciuti, Silvia
Babich, Francesco
Climate change is driving the construction sector to use of more environmentally friendly and sustainable materials. Hemp concrete has been recently adopted as an innovative solution by the building industry to reduce emissions, as this material stores more CO2 than the emitted during its production. Part of this storage occurs during its service life leading to a reduction of indoor CO2 levels. CO2 has been widely used as a proxy for evaluating indoor air quality (IAQ). However, these assessments do not consider the features of carbon sequestration materials where the use of CO2 as the sole indicator might lead to a poor IAQ. This paper describes the results of an IAQ monitoring campaign conducted in a CO2-negative prototype house built with hemp concrete and wood. During the campaign, temperature, relative humidity, total volatile organic compounds (TVOC), and carbon dioxide (CO2) levels were continuously monitored inside and outside the test house for three months. Due to the low occupancy of the prototype house, the indoor CO2 levels were generally low. These levels ranged between 400 and 1100 ppm when people was present and decreased down to 30 ppm once the visitors were gone and the house was closed. Such low indoor CO2 levels suggest that hemp concrete walls could sequestrate a significant amount of CO2. During the same period, TVOC concentration varied between 0.19 and 3.62 ppm and was negatively correlated to CO2 levels. The highest TVOC concentrations were recorded in the absence of visitors when the prototype house was closed, indicating that building materials or furniture could potentially emit such compounds. The comparison of CO2 and TVOC levels showed that, while the CO2 levels were very low, the levels of TVOC were at their maximum. Consequently, in buildings made of hemp concrete, the use of CO2 as a single indicator for ventilation might be misleading and could result in the occupants exposure to high pollution levels. Despite the importance of CO2 monitoring, the findings of this study indicate that in hemp concrete buildings additional measurements may be needed to assess IAQ and identify potential sources of pollution. In some cases, adapting ventilation strategies to CO2 levels could be sufficient to ensure acceptable IAQ, however, a preliminary assessment of IAQ should be conducted before making any decisions.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/373327
info:eu-repo/semantics/altIdentifier/isbn/9782930471631
https://www.aivc.org/resource/hemp-concrete-walls-evaluation-relationship-between-co2-and-tvoc
eng
ispartofbook:AIVCProceedings 42nd AIVC-10th TightVent - 8th venticool Conference
42nd AIVC - 10th TightVent - 8th venticool Conference
numberofpages:10
info:eu-repo/semantics/openAccess
AIVC
place:Rotterdam, Netherlands
oai:iris.unitn.it:11572/3913892024-01-17T00:11:26Zcom_11572_101873com_11572_101871col_11572_101864
Complex Daily Activities, Country-Level Diversity, and Smartphone Sensing: A Study in Denmark, Italy, Mongolia, Paraguay, and UK
Assi K.
Meegahapola L.
Droz W.
Kun P.
De Gotzen A.
Bidoglia M.
Stares S.
Gaskell G.
Chagnaa A.
Ganbold A.
Zundui T.
Caprini C.
Miorandi D.
Zarza J. L.
Hume A.
Cernuzzi L.
Bison I.
Rodas Britez M. D.
Busso M.
Chenu-Abente R.
Giunchiglia F.
Gatica-Perez D.
Assi, K.
Meegahapola, L.
Droz, W.
Kun, P.
De Gotzen, A.
Bidoglia, M.
Stares, S.
Gaskell, G.
Chagnaa, A.
Ganbold, A.
Zundui, T.
Caprini, C.
Miorandi, D.
Zarza, J. L.
Hume, A.
Cernuzzi, L.
Bison, I.
Rodas Britez, M. D.
Busso, M.
Chenu-Abente, R.
Giunchiglia, F.
Gatica-Perez, D.
smartphone sensing, complex activities of daily living, model generalization, behavior recognition, diversity-awareness, passive sensing, domain shift, context-awareness, activity recognition, distributional shift
Smartphones enable understanding human behavior with activity recognition to support people’s daily lives. Prior studies focused on using inertial sensors to detect simple activities (sitting, walking, running, etc.) and were mostly conducted in homogeneous populations within a country. However, people are more sedentary in the post-pandemic world with the prevalence of remote/hybrid work/study settings, making detecting simple activities less meaningful for context-aware applications. Hence, the understanding of (i) how multimodal smartphone sensors and machine learning models could be used to detect complex daily activities that can better inform about people’s daily lives, and (ii) how models generalize to unseen countries, is limited. We analyzed in-the-wild smartphone data and ∼ 216K self-reports from 637 college students in five countries (Italy, Mongolia, UK, Denmark, Paraguay). Then, we defined a 12-class complex daily activity recognition task and evaluated the performance with different approaches. We found that even though the generic multi-country approach provided an AUROC of 0.70, the country-specific approach performed better with AUROC scores in [0.79-0.89]. We believe that research along the lines of diversity awareness is fundamental for advancing human behavior understanding through smartphones and machine learning, for more real-world utility across countries.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/391389
10.1145/3544548.3581190
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85160011386
info:eu-repo/semantics/altIdentifier/isbn/978-1-4503-9421-5
https://dl.acm.org/doi/10.1145/3544548.3581190
info:eu-repo/semantics/altIdentifier/wos/WOS:001048393801019
eng
ispartofbook:Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems
CHI '23
firstpage:1
lastpage:23
numberofpages:23
info:eu-repo/semantics/closedAccess
Association for Computing Machinery
place:New York, NY, USA
oai:iris.unitn.it:11572/3558212024-03-21T21:01:19Zcom_11572_101873com_11572_101871col_11572_101864
Physical Layer Authentication in Underwater Acoustic Networks with Mobile Devices
Paolo Casari
Francesco Ardizzon
Stefano Tomasin
Casari, Paolo
Ardizzon, Francesco
Tomasin, Stefano
Underwater acoustic networks, mobility, authentication, physical layer security, simulation
As underwater acoustic communication technologies become mature and underwater networks evolve into reliable solutions for data communications, authenticating transmitted data turns from an option to a necessity.
In this paper, we explore physical-layer authentication for an underwater acoustic networks with mobile devices. We choose the power-weighted average of the channel taps' arrival delay as the main authentication feature. We then develop a Kalman filter approach to track the evolution of this feature in the presence of mobility. The filter computes an innovation metric for each new transmission, which is processed to determine if a signal originates from a legitimate network node. Simulation results obtained from a dataset generated with Bellhop show that our authentication mechanism successfully distinguishes between legitimate and impersonating transmitters. Moreover, we show that linearly combining innovation readings from multiple sensors yields a good low-complexity classifier, and assess the impact of the transmitter speed on the authentication performance.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/355821
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85145881072
info:eu-repo/semantics/altIdentifier/isbn/978-1-4503-9952-4
https://dl.acm.org/doi/10.1145/3567600.3567604
eng
ispartofbook:Proc. of the 16th International Conference on Underwater Networks & Systems (WUWNet'22)
WUWNet
info:eu-repo/semantics/openAccess
ELETTRONICO
ACM
place:United States
oai:iris.unitn.it:11572/3519422024-03-22T22:42:49Zcom_11572_101873com_11572_101871col_11572_101864
Interpretable AI for policy-making in pandemics
Custode, Leonardo Lucio
Iacca, Giovanni
Fieldsend, Jonathan E.
Custode, Leonardo Lucio
Iacca, Giovanni
Since the first wave of the COVID-19 pandemic, governments have applied restrictions in order to slow down its spreading. However, creating such policies is hard, especially because the government needs to trade-off the spreading of the pandemic with the economic losses. For this reason, several works have applied machine learning techniques, often with the help of special-purpose simulators, to generate policies that were more effective than the ones obtained by governments. While the performance of such approaches are promising, they suffer from a fundamental issue: since such approaches are based on black-box machine learning, their real-world applicability is limited, because these policies cannot be analyzed, nor tested, and thus they are not trustable. In this work, we employ a recently developed hybrid approach, which combines reinforcement learning with evolutionary computation, for the generation of interpretable policies for containing the pandemic. These policies, trained on an existing simulator, aim to reduce the spreading of the pandemic while minimizing the economic losses. Our results show that our approach is able to find solutions that are extremely simple, yet very powerful. In fact, our approach has significantly better performance (in simulated scenarios) than both previous work and government policies.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/351942
10.1145/3520304.3533959
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85136333383
info:eu-repo/semantics/altIdentifier/isbn/9781450392686
https://dl.acm.org/doi/10.1145/3520304.3533959
info:eu-repo/semantics/altIdentifier/wos/WOS:001035469400277
eng
ispartofbook:GECCO '22: Proceedings of the Genetic and Evolutionary Computation Conference Companion
GECCO '22
firstpage:1763
lastpage:1769
numberofpages:7
info:eu-repo/semantics/openAccess
alleditors:Fieldsend, Jonathan E.
ACM
place:New York
oai:iris.unitn.it:11572/3580072023-12-02T23:36:39Zcom_11572_101873com_11572_101871col_11572_101864
Holomorphic functions and regular quaternionic functions on the hyperkähler space H
Perotti, Alessandro
Perotti, Alessandro
Quaternionic regular functions, hyperk\ahler manifolds"
Let H be the space of quaternions, with its standard hypercomplex structure. Let $\mathcal R(\Omega)$ be the module of \emph{$\psi$-regular} functions on $\Omega$. For every $p\in H$, $p^2=-1$, $\mathcal R(\Omega)$ contains the space of holomorphic functions w.r.t. the complex structure $J_p$ induced by $p$. We prove the existence, on any bounded domain $\Omega$, of $\psi$-regular functions that are not $J_p$-holomorphic for any $p$. Our starting point is a result of Chen and Li concerning maps between hyperk\ahler manifolds
2005
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/358007
eng
ispartofbook:[s.n.]
V ISAAC Congress
firstpage:1
lastpage:9
numberofpages:9
info:eu-repo/semantics/openAccess
ELETTRONICO
World Scientific
place:[s.l.]
oai:iris.unitn.it:11572/3748322024-02-10T04:57:00Zcom_11572_101873com_11572_101871col_11572_101864
Barbara Sokołowska Brukalski. Graphic Analysis of the House on Niegolewskiego Street
Vattano, Starlight
Castaño Perea, Enrique
Echeverría Valiente, Ernesto
Vattano, Starlight
Representation, Graphic analysis, Barbara Sokołowska Brukalski, Digital modeling, History of Architecture
The project analyzed in this paper by the critical and digital redrawing is the House on Niegolewskiego Street of 1927–28, designed by Barbara Sokołowska for her and her husband, Stanislaw Brukalski, inspired by the Neoplasticism ideas, the paintings of Mondrian and De Stijl. In its details the house seems to recall that one in Utrecht by G.T. Rietveld of 1924, with a very large living room, a studio overlooking a double height and a spiral stair connecting the two floors into one space defined by translated planes and pure volumes.
2018
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/374832
10.1007/978-3-319-58856-8_96
info:eu-repo/semantics/altIdentifier/isbn/978-3-319-58855-1
https://link.springer.com/chapter/10.1007/978-3-319-58856-8_96
eng
ispartofbook:Architectural Draughtsmanship. From Analog to Digital Narratives
16th International Congress EGA (Architectual Graphic Expression). Teaching and researching in architectural graphic expression.
firstpage:1221
lastpage:1231
numberofpages:11
info:eu-repo/semantics/openAccess
ELETTRONICO
alleditors:Castaño Perea, Enrique; Echeverría Valiente, Ernesto
Springer
country:CHE
place:Cham
oai:iris.unitn.it:11572/4008712024-03-24T07:55:40Zcom_11572_101873com_11572_101871col_11572_101864
A Simple Latent Variable Model for Graph Learning and Inference
Jaeger, Manfred
Longa, Antonio
Azzolin, Steve
Schulte, Oliver
Passerini, Andrea
Jaeger, Manfred
Longa, Antonio
Azzolin, Steve
Schulte, Oliver
Passerini, Andrea
We introduce a probabilistic latent variable model for graphs that generalizes
both the established graphon and stochastic block models. This naive histogram
AHK model is simple and versatile, and we demonstrate its use for disparate
tasks including complex predictive inference usually not supported by other
approaches, and graph generation. We analyze the tradeoffs entailed by the
simplicity of the model, which imposes certain limitations on expressivity on
the one hand, but on the other hand leads to robust generalization capabilities to
graph sizes different from what was seen in the training data.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/400871
https://openreview.net/forum?id=S9jem2KZVr
eng
ispartofbook:Proceedings of the Second Learning on Graphs Conference (LoG 2023)
LOG 2023
volume:231
firstpage:1
lastpage:18
numberofpages:18
serie:PROCEEDINGS OF MACHINE LEARNING RESEARCH
info:eu-repo/semantics/openAccess
ML Research Press
place:New York
oai:iris.unitn.it:11572/3604832024-03-23T00:58:16Zcom_11572_101873com_11572_101871col_11572_101864
Learning Atrial Fiber Orientations and Conductivity Tensors from Intracardiac Maps Using Physics-Informed Neural Networks
Grandits T.
Pezzuto S.
Costabal F. S.
Perdikaris P.
Pock T.
Plank G.
Krause R.
Ennis, Daniel B
Perotti, Luigi E.
Wang, Vicky Y.
Grandits, T.
Pezzuto, S.
Costabal, F. S.
Perdikaris, P.
Pock, T.
Plank, G.
Krause, R.
Electroanatomical maps are a key tool in the diagnosis and treatment of atrial fibrillation. Current approaches focus on the activation times recorded. However, more information can be extracted from the available data. The fibers in cardiac tissue conduct the electrical wave faster, and their direction could be inferred from activation times. In this work, we employ a recently developed approach, called physics informed neural networks, to learn the fiber orientations from electroanatomical maps, taking into account the physics of the electrical wave propagation. In particular, we train the neural network to weakly satisfy the anisotropic eikonal equation and to predict the measured activation times. We use a local basis for the anisotropic conductivity tensor, which encodes the fiber orientation. The methodology is tested both in a synthetic example and for patient data. Our approach shows good agreement in both cases, with an RMSE of 2.2 ms on the in-silico data and outperforming a state of the art method on the patient data. The results show a first step towards learning the fiber orientations from electroanatomical maps with physics-informed neural networks.
2021
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/360483
10.1007/978-3-030-78710-3_62
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85111807691
info:eu-repo/semantics/altIdentifier/pmid/35098259
eng
ispartofbook:Functional Imaging and Modeling of the Heart 11th International Conference, FIMH 2021
FIMH2021
volume:12738
firstpage:650
lastpage:658
numberofpages:9
serie:LECTURE NOTES IN COMPUTER SCIENCE
STAMPA
alleditors:Ennis, Daniel B; Perotti, Luigi E.; Wang, Vicky Y.
Springer Cham
place:Stanford
oai:iris.unitn.it:11572/3696602023-05-09T12:58:22Zcom_11572_101873com_11572_101871col_11572_101864
A Diversity-Aware Domain Development Methodology
Bagchi, Mayukh
Bagchi, Mayukh
Representation Diversity, Ontology Concept Reuse, Teleology, Domain Development Methodology
The development of domain ontological models, though being a mature research arena backed by wellestablished methodologies, still suffer from two key shortcomings. Firstly, the issues concerning the
semantic persistency of ontology concepts and their flexible reuse in domain development employing
existing approaches. Secondly, due to the difficulty in understanding and reusing top-level concepts in
existing foundational ontologies, the obfuscation regarding the semantic nature of domain representations.
The paper grounds the aforementioned shortcomings in representation diversity and proposes a three-fold
solution - (i) a pipeline for rendering concepts reuse-ready, (ii) a first characterization of a minimalistic
foundational knowledge model, named foundational teleology, semantically explicating foundational
distinctions enforcing the static as well as dynamic nature of domain representations, and (iii) a flexible,
reuse-native methodology for diversity-aware domain development exploiting solutions (i) and (ii). The
preliminary work reported validates the potentiality of the solution components.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/369660
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85138712811
https://ceur-ws.org/Vol-3211/
eng
ispartofbook:Proceedings of the ER Forum and PhD Symposium 2022
ER
firstpage:1
lastpage:11
numberofpages:11
serie:CEUR WORKSHOP PROCEEDINGS
info:eu-repo/semantics/openAccess
CEUR
place:Aachen
oai:iris.unitn.it:11572/3770082024-03-17T21:57:23Zcom_11572_101873com_11572_101871col_11572_101864
On the Feasibility of RIS-enabled Cooperative Driving
Segata, Michele
Casari, Paolo
Lestas, Marios
Tyrovolas, Dimitrios
Saeed, Taqwa
Karagiannidis, George K.
Liaskos, Christos
Segata, Michele
Casari, Paolo
Lestas, Mario
Tyrovolas, Dimitrio
Saeed, Taqwa
Karagiannidis, George K.
Liaskos, Christos
Wireless communication, Target tracking, Transmitters, Instruments, Ecosystems, Focusing, Receivers
Future cooperative autonomous vehicles will require
high-performance communication means to support functions
such as cooperative maneuvering and cooperative perception.
The high-bandwidth requirements of these functions can be
met through mmWave communications, whose utilization is
often hindered by the harsh propagation conditions of typical
vehicular environments. A solution to this problem is the use
of reconfigurable intelligent surfaces (RISs), which enable the
reflection of signals in a configurable direction, and have recently
gained attention in the vehicular domain. In this paper, we
provide an initial feasibility study, highlighting the challenges
ahead and the performance RISs need to deliver in order
to enable this type of communications. Specifically, we utilize
CoopeRIS, a simulation framework for RISs integrated into
the Plexe/Veins/SUMO ecosystem that we develop as further
contribution and will release to the public.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/377008
10.1109/VNC57357.2023.10136324
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-3549-1
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-3550-7
https://ieeexplore-ieee-org.ezp.biblio.unitn.it/document/10136324
info:eu-repo/semantics/altIdentifier/wos/WOS:001011821500022
eng
ispartofbook:14th IEEE Vehicular Networking Conference (VNC 2023)
IEEE Vehicular Networking Conference
firstpage:143
lastpage:150
numberofpages:8
serie:Vehicular Networking Conference (VNC)
info:eu-repo/semantics/openAccess
IEEE
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3725062023-04-06T11:28:39Zcom_11572_101873com_11572_101871col_11572_101864
Radiomics-based non-invasive lymph node metastases prediction in breast cancer
Cordelli E
Sicilia R
Santucci D
De Felice C
Quattrocchi CC
Beomonte Zobel B
Iannello G
Soda P
de Herrera AGS. Rodriguez Gonzalez A Santosh KC Temesgen Z Kane B Soda P
Cordelli, E
Sicilia, R
Santucci, D
De Felice, C
Quattrocchi, Cc
Beomonte Zobel, B
Iannello, G
Soda, P
Breast cancer is the most common tumour in women and it is characterized by a huge variety of clinical and histological scenarios and imaging pattern. The axillary lymph node metastases presence or absence is one of the most important prognostic factors affecting the loco-regional recurrence and the overall survival. The lymph node status is usually determined by an histological exam, an invasive procedure that could result in complications. This work aims to provide a safer and non-invasive prognostic approach by introducing a radiomics-based method that predicts axillary lymph node metastasis. It combines primary tumor histological features and patients clinical data with quantitative measures extracted from the MR images. To compute these latter quantities we determine the convex hull of the ROIs and we introduce the Three Orthogonal Planes-Local Binary Pattern (TOP-LBP). On 99 samples the approach achieves a promising AUC equal to 85.6%.
2020
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/372506
10.1109/CBMS49503.2020.00098
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85091143695
info:eu-repo/semantics/altIdentifier/wos/WOS:000786468800090
eng
ispartofbook:2020 IEEE 33rd International Symposium on Computer-Based Medical Systems (CBMS),
IEEE Symposium on Computer-Based Medical Systems
firstpage:486
lastpage:491
numberofpages:6
info:eu-repo/semantics/closedAccess
alleditors:de Herrera AGS. Rodriguez Gonzalez A Santosh KC Temesgen Z Kane B Soda P
IEEE Computer Society
place:Los Alamitos, California
oai:iris.unitn.it:11572/3950892024-03-05T02:05:21Zcom_11572_101873com_11572_101871col_11572_101864
Attribute-Preserving Face Dataset Anonymization via Latent Code Optimization
Barattin, Simone
Tzelepis, Christos
Patras, Ioannis
Sebe, Nicu
Barattin, Simone
Tzelepis, Christo
Patras, Ioanni
Sebe, Nicu
Training, Data privacy, Codes, Generative adversarial networks, Information filtering,
Pattern recognition, Task analysis
This work addresses the problem of anonymizing the identity of faces in a dataset of images, such that the privacy of those depicted is not violated, while at the same time the dataset is useful for downstream task such as for training machine learning models. To the best of our knowledge, we are the first to explicitly address this issue and deal with two major drawbacks of the existing state-of-the-art approaches, namely that they (i) require the costly training of additional, purpose-trained neural networks, and/or (ii) fail to retain the facial attributes of the original images in the anonymized counterparts, the preservation of which is of paramount importance for their use in downstream tasks. We accordingly present a task-agnostic anonymization procedure that directly optimizes the images' latent representation in the latent space of a pretrained GAN. By optimizing the latent codes directly, we ensure both that the identity is of a desired distance away from the original (with an identity obfuscation loss), whilst preserving the facial attributes (using a novel feature-matching loss in FaRL's [48] deep feature space). We demonstrate through a series of both qualitative and quantitative experiments that our method is capable of anonymizing the identity of the images whilst-crucially-better-preserving the facial attributes. We make the code and the pretrained models publicly available at: https://github.com/chi0tzp/FALCO.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/395089
10.1109/CVPR52729.2023.00773
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85173971376
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-0129-8
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-0130-4
https://ieeexplore.ieee.org/document/10203624
info:eu-repo/semantics/altIdentifier/wos/WOS:001062522100001
eng
ispartofbook:2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
CVPR
firstpage:8001
lastpage:8010
numberofpages:10
serie:PROCEEDINGS IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION
info:eu-repo/semantics/openAccess
IEEE
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3599052024-03-16T01:08:41Zcom_11572_101873com_11572_101871col_11572_101864
“Un grand nombre de productions des maîtres les plus célèbres, ignorées depuis longues années”: una panoramica sui dipinti rinascimentali della Recueil di Jean-Baptiste Pierre Lebrun
Mattedi, Luca
Ferrari, Stefano
Mattedi, Luca ... [et al.]
Miarelli Mariani, I.
Casola, T.
Fraticelli, V.
Lisanti, V.
Palombaro, L.
Mattedi, Luca
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/359905
info:eu-repo/semantics/altIdentifier/isbn/978-88-85795-83-9
http://www.campisanoeditore.it/index.php/10-collana-storia-dell-arte/322-la-storia-dell-arte-illustrata
ita
ispartofbook:La storia dell’arte illustrata e la stampa di traduzione tra XVIII e XIX secolo
La storia dell’arte illustrata e la stampa di traduzione tra XVIII e XIX secolo
firstpage:85
lastpage:94
numberofpages:10
info:eu-repo/semantics/closedAccess
alleditors:Miarelli Mariani, I.; Casola, T.; Fraticelli, V.; Lisanti, V.; Palombaro, L.
Campisano Editore Srl
country:ITA
place:Roma
oai:iris.unitn.it:11572/4015922024-02-09T02:38:34Zcom_11572_101873com_11572_101871col_11572_101864
Ecosystem of clusters made of self-propelled droplet surfers
Löffler, Richard J. G.
Tanaka, Shinpei
Holler, Silvia
Hanczyc, Martin M.
Hiroyuki Iizuka, Keisuke Suzuki, Ryoko Uno, Luisa Damiano, Nadine Spychala, Miguel Aguilera, Eduardo J. Izquierdo, Reiji Suzuki and Manuel Baltieri
Löffler, Richard J. G.
Tanaka, Shinpei
Holler, Silvia
Hanczyc, Martin M.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/401592
10.1162/isal_a_00587
https://direct.mit.edu/isal/proceedings/isal/35/130/116849
eng
ispartofbook:ALIFE 2023: Ghost in the Machine: Proceedings of the 2023 Artificial Life Conference
ALIFE 2023
numberofpages:3
serie:ALIFE: PROCEEDINGS OF THE ARTIFICIAL LIFE CONFERENCE
info:eu-repo/semantics/openAccess
ELETTRONICO
alleditors:Hiroyuki Iizuka, Keisuke Suzuki, Ryoko Uno, Luisa Damiano, Nadine Spychala, Miguel Aguilera, Eduardo J. Izquierdo, Reiji Suzuki and Manuel Baltieri
MIT Press
country:USA
place:Cambridge, Massachusetts USA
oai:iris.unitn.it:11572/4032992024-03-03T11:13:56Zcom_11572_101873com_11572_101871col_11572_101864
Evaluating Deep Neural Networks for Face Recognition with Different Subsets of Mugshots From the Photo-Signaling Procedure
Contardo, Paolo
Rossini, Nicolò
Tomassini, Selene
Falcionelli, Nicola
Dragoni, Aldo Franco
Sernani, Paolo
Contardo, Paolo
Rossini, Nicolò
Tomassini, Selene
Falcionelli, Nicola
Dragoni, Aldo Franco
Sernani, Paolo
Face Recognition, Crime Investigation, Face Verification, Face Identification, Law Enforcement
Whereas Face Recognition with Convolutional Neural Networks (CNNs) is considered a mature technology to the point that, in addition to applications such as smartphone unlocking and passport verification, it is used in already existing decision support systems for crime investigations, there still are open challenges in Pose-Invariant Face Recognition (PIFR). Specifically, there is a lack of research in understanding how subsets of mugshots different from the frontal and right profile pictures routinely collected by police forces during the photo-signaling procedure might impact on the Face Recognition accuracy in security camera videos recorded “in the wild”. To this end, we compare two well-known CNNs for Face Recognition, namely VGG16 and ResNet50, on the Face Recognition from Mugshots DataBase (FRMDB), specifically designed to evaluate the performance in Face Recognition using different subsets of mugshots. With respect to our previous research, we collect more general results, testing with more identities on additional security camera videos.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/403299
10.1109/MetroXRAINE58569.2023.10405736
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-0080-2
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-0081-9
https://ieeexplore.ieee.org/document/10405736
eng
ispartofbook:2023 IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE)
2nd IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering, MetroXRAINE 2023
firstpage:543
lastpage:548
numberofpages:6
info:eu-repo/semantics/closedAccess
ELETTRONICO
Institute of Electrical and Electronics Engineers (IEEE)
place:New York City; Piscataway, New Jersey
oai:iris.unitn.it:11572/3704122023-06-16T08:18:16Zcom_11572_101873com_11572_101871col_11572_101864
Reverse-engineering the language of thought: A new approach
Denic M.
Szymanik J.
Denic, M.
Szymanik, J.
language of thought
numeral
number
complexity/informativeness trade-off
A foundational hypothesis in cognitive science is that some
of human thinking happens in a language of thought (LoT),
which is universal across humans (Fodor, 1975). According
to this hypothesis, words in different natural languages are
labels for primitive concepts or their combinations in LoT.
What are LoT’s primitives? This is a major challenge because
LoT is not directly observable, and thus needs to be inferred
or reverse-engineered. We put forward a novel approach to
reverse-engineering LoT, capitalizing on the existing knowledge about the optimization of the trade-off between complexity and informativeness in natural languages.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/370412
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85146422538
https://escholarship.org/uc/item/3kw4p77n
eng
ispartofbook:Proceedings of the 44th Annual Meeting of the Cognitive Science Society
CogSci
volume:44
firstpage:2918
lastpage:2925
numberofpages:8
info:eu-repo/semantics/openAccess
eScholarship Publishing
country:USA
place:California (USA)
oai:iris.unitn.it:11572/3909902023-10-06T22:55:04Zcom_11572_101873com_11572_101871col_11572_101864
An Intuitive Augmented Reality Interface for Task Scheduling, Monitoring, and Work Performance Improvement in Human-Robot Collaboration
De Franco, Alessandro
Lamon, Edoardo
Balatti, Pietro
De Momi, Elena
Ajoudani, Arash
De Franco, Alessandro
Lamon, Edoardo
Balatti, Pietro
De Momi, Elena
Ajoudani, Arash
2019
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/390990
10.1109/IWOBI47054.2019.9114472
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85087282242
eng
ispartofbook:2019 IEEE International Work Conference on Bioinspired Intelligence (IWOBI)
IWOBI
firstpage:75
lastpage:80
numberofpages:6
IEEE
place:Piscataway, NJ
oai:iris.unitn.it:11572/3556962024-03-10T08:42:41Zcom_11572_101873com_11572_101871col_11572_101864
Public engagement and gender differences in Italy: Exploring the gap by activity and discipline
Anzivino, Monia
Anzivino, Monia
Public engagement, Gender differences, Science in society
The extent of some changes that have occurred over the last years, as well as the social and political consequences of these changes, has led to the emergence of new needs related to expert knowledge, scientific culture, and public trust, calling for new attention regarding the implications of the ties between science and society. Scientists are called upon in first person to engage with society, interact and facilitate the relationship between the scientific world and the world of non-experts. This study focuses on the public engagement activities of Italian academics, specifically on the participation of women scientists in these activities since the relationship between gender and public engagement in the literature is unclear. What is the involvement of women in these activities? Is it different in terms of quantity and quality from that of men? In which disciplines the gender gap, if any, is stronger? How do attitudes towards the university’s role in society impact on public engagement of men and women? Using survey data from a national sample of Italian academics from all disciplines (N=5.123), we find that men and women are equally engaged in community-based activities, but women are less engaged in communication activities through mass media. Moreover, the gender gap in the last group of activities is stronger in those disciplines where academics are more frequently engaged, Health Sciences among STEM disciplines and Social Sciences among SSH. These results suggest a different analytical approach for investigating gender differences in public engagement and indicate which disciplinary fields need more incisive policies for promoting women as experts.
2021
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/355696
info:eu-repo/semantics/altIdentifier/isbn/978-88-944888-7-6
https://www.scuolademocratica-conference.net/proceedings-2/
eng
ispartofbook:Proceedings of the 2nd International Conference of the Journal Scuola Democratica: Reinventing Education, vol. 3: Pandemic and Post-Pandemic Space and Time
2nd International Conference of the Journal Scuola Democratica
firstpage:285
lastpage:296
numberofpages:12
info:eu-repo/semantics/openAccess
Associazione "Per Scuola Democratica"
country:ITA
place:Roma
oai:iris.unitn.it:11572/4016632024-02-13T01:14:12Zcom_11572_101873com_11572_101871col_11572_101864
An Analytical Approach for Multi-Panel CLT Shearwalls Attached to the Base with Equally Spaced Anchors
Masroor M.
Casagrande D.
Doudak G.
Masroor, M.
Casagrande, D.
Doudak, G.
Analytical approach
Cross-Laminated Timber
Elastic
Inelastic
Multi-panel shearwall
Uniform screw
Vertical joints
The ability of Cross-Laminated Timber (CLT) walls to resist gravity loads has been well-documented, and the required design provisions have been established. CLT shearwalls have also been successfully used as lateral load resistance systems. Although several analytical approaches have been developed to establish the mechanical behaviour of CLT shearwalls, no published methodology has been found that covers the case of multi-panel CLT shearwalls anchored directly to the floor using screws instead of mechanical anchors. An elastic methodology, developed using the minimum potential energy method, has been established to enable analysis of multi-panel CLT shearwalls connected to the foundation or floor below and floor above using equally spaced anchors. Screw fasteners are used to demonstrate the proposed concept. The developed elastic method is presented for a two-panel CLT shearwall to also incorporate the inelastic properties of the connections. The consistency kinematic regions are plotted to demonstrate how different wall configurations affect the final lateral mode of the shearwalls. The elastic capacity of the developed multi-panel shearwalls is established. The analytical expressions are verified by comparing them with numerical models with near perfect match observed.
2021
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/401663
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85120754471
info:eu-repo/semantics/altIdentifier/isbn/9781713841111
eng
ispartofbook:World Conference on Timber Engineering 2021, WCTE 2021
World Conference on Timber Engineering 2021, WCTE 2021
info:eu-repo/semantics/closedAccess
Curran Associates, Inc. ( Jan 2022 )
place:Santiago, Chile
oai:iris.unitn.it:11572/3526812024-03-22T21:50:30Zcom_11572_101873com_11572_101871col_11572_101864
Multimodal Emotion Recognition with Modality-Pairwise Unsupervised Contrastive Loss
Franceschini, Riccardo
Fini Enrico
Beyan Cigdem
Conti, Alessandro
Arrigoni, Federica
Ricci, Elisa
Franceschini, Riccardo
Fini, Enrico
Beyan, Cigdem
Conti, Alessandro
Arrigoni, Federica
Ricci, Elisa
Emotion recognition is involved in several real-world applications. With an increase in available modalities, automatic understanding of emotions is being performed more accurately. The success in Multimodal Emotion Recognition (MER), primarily relies on the supervised learning paradigm. However, data annotation is expensive, time-consuming, and as emotion expression and perception depends on several factors (e.g., age, gender, culture) obtaining labels with a high reliability is hard. Motivated by these, we focus on unsupervised feature learning for MER. We consider discrete emotions, and as modalities text, audio and vision are used. Our method, as being based on contrastive loss between pairwise modalities, is the first attempt in MER literature. Our end-to-end feature learning approach has several differences (and advantages) compared to existing MER methods: i) it is unsupervised, so the learning is lack of data labelling cost; ii) it does not require data spatial augmentation, modality alignment, large number of batch size or epochs; iii) it applies data fusion only at inference; and iv) it does not require backbones pre-trained on emotion recognition task. The experiments on benchmark datasets show that our method outperforms several baseline approaches and unsupervised learning methods applied in MER. Particularly, it even surpasses a few supervised MER state-of-the-art.
2022
info:eu-repo/semantics/conferenceObject
http://hdl.handle.net/11572/352681
10.1109/ICPR56361.2022.9956589
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-9062-7
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-9063-4
https://www.icpr2022.com/
info:eu-repo/semantics/altIdentifier/wos/WOS:000897707602082
eng
ispartofbook:26st International Conference on Pattern Recognition (ICPR) 2022
ICPR
firstpage:2589
lastpage:2596
numberofpages:8
serie:INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION
info:eu-repo/grantAgreement/EC/H2020/871245
info:eu-repo/semantics/closedAccess
ELETTRONICO
IEEE/ IAPR
place:New York, USA
oai:iris.unitn.it:11572/3586032023-12-02T23:37:52Zcom_11572_101873com_11572_101871col_11572_101864
Britain Re-Creates the Social Movement: Contentious (and Not-so-contentious) Networks in Glasgow
Diani, Mario
Diani, Mario
social movement
social network
UK politics
Looking at networks of civic organizations in Glasgow, I try to single out social movement dynamics from a broader set of collaborative interactions and shared memberships between organizations mobilizing on environmental, ethnic and minority, and social exclusion issues. In doing so, I also draw attention to the persistent usefulness of the concept of social movement, which recent work by Tilly (McAdam et al., 2001) seems to have placed in a less central position.
2002
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/358603
eng
ispartofbook:-
Contentious Politics and the Economic Opportunity Structure
numberofpages:27
info:eu-repo/semantics/openAccess
ELETTRONICO
s.n.]
place:[s.l.
oai:iris.unitn.it:11572/3624252024-03-22T18:47:24Zcom_11572_101873com_11572_101871col_11572_101864
Is NLP-based Test Automation Cheaper Than Programmable and Capture &Replay?
Maurizio, Leotta
Filippo, Ricca
Simone, Stoppa
Alessandro, Marchetto
Vallecillo, A.
Visser, J.
Pérez-Castillo, R.
Leotta, Maurizio
Ricca, Filippo
Stoppa, Simone
Marchetto, Alessandro
Test Automation · Web Testing · NLP · Artificial Intelligence.
Nowadays, there is a growing interest in the use of Natural-Language Processing (NLP) for supporting software test automation. This paper investigates the adoption of NLP in web testing. To this aim, a case study has been conducted to compare the cost of the adoption of a NLP testing approach, with respect to more consolidated approaches, i.e., programmable testing and capture and replay testing, in two testing tasks: test cases development and test case evolution/maintenance.
Even if preliminary, results show that NLP testing is quite competitive with respect to the more consolidated approaches since the cumulative testing effort of a NLP testing approach, computed considering both development and evolution efforts, is almost always lower than the one of programmable testing and capture&replay testing.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/362425
10.1007/978-3-031-14179-9_6
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85137976205
info:eu-repo/semantics/altIdentifier/isbn/978-3-031-14178-2
https://link.springer.com/chapter/10.1007/978-3-031-14179-9_6#citeas
info:eu-repo/semantics/altIdentifier/wos/WOS:000870625200006
eng
ispartofbook:Quality of Information and Communications Technology. QUATIC 2022.
15th International Conference on Quality of Information and Communications Technology, QUATIC 2022
volume:1621
firstpage:77
lastpage:92
numberofpages:16
serie:COMMUNICATIONS IN COMPUTER AND INFORMATION SCIENCE
info:eu-repo/semantics/embargoedAccess
alleditors:Vallecillo, A.; Visser, J.; Pérez-Castillo, R.
Springer
country:USA
place:Cham, Svizzera
oai:iris.unitn.it:11572/3748372024-02-11T01:11:12Zcom_11572_101873com_11572_101871col_11572_101864
L'attualità della pedagogia delle arti di inizio Novecento in Est Europa: Gestaltung e Formenlehre come risorse delle creatività figurative
Luigini, Alessandro
Vattano, Starlight
Marcos Alba, Carlos Lui
Juan Gutiérrez, Pablo Jeremía
Domingo Gresa, Jorge
Oliva Meyer, Justo
Luigini, Alessandro
Vattano, Starlight
Gestaltung, Proun, Formenlehre, didattica delle creatività figurative = Gestaltung, Proun, Formenlehre, teaching of figuratives art
In the last century the great innovations in the field of the forms and figures elaboration have often sometimes accompanied the amazement and optimism of the discovery with a methodological uncertainty shared with other fields of Knowledge. It is undeniable that the advent of information technology and representation - together with the socio-economic changes that have taken place - has laid the foundations for an irreversible ontological leap in the process of form processing, and yet very often there are historical references that can somehow inspire current theories and practices. In this sense, some interpretations considered to be classical, from Valéry to Benjamin, from Arnheim to Waburg, provide not a few substantial points for understanding contemporaneity. In this regard, the well-known pedagogical sketches by Paul Klee (1925), like the whole compendium of notes that the Swiss artist wrote and cataloged from the first day of teaching at the Bauhaus (1921-31), contain ideas and practices that can be declined in the world of graphic informatics, parametric modeling software, but also in areas where the goal is not the control of the form as such, to the advantage of the creative process control. And a renewed interest in the work of Klee pedagogue is also due to the open source accessibility guaranteed by the Zentrum Paul Klee of Bremen, which allows a direct study of the sources. Another extraordinarily interesting idea is El Lisintskij’s cross-media production, who with his Proun anticipates - or inspires - the work of some of the greatest architects of the late twentieth century. El Lisintskij’s work is also characterized by the free overflow of visual arts - graphics, painting, sculpture and architecture - of compositional methods, production of stylistic elements and figurative innovations, laying the foundations for a vision of art that we could define as “total”. Other authors of the first two decades of the twentieth century have explored the potential of the renewed figurative arts: from the Kandinsky’s lucky “Point, line, surface” to the Theory of form and composition by Itten, the pedagogy of the Bauhaus arts is, evidently, still valid as a cultural substratum for research in the architectural field, but perhaps more generally, in the figurative arts. But if the pedagogical structure of the Bauhaus and the developments of Russian constructivism are well traceable in literature, just from the exchange with these two great avant-garde experiences, all a graphic-artistic-performative production in the East-European area seems to have developed and hybridized compositional themes permuted by Bauhaus and Russian Constructivism.
Le grandi innovazioni nel campo dell’elaborazione di forme e figure nel secolo scorso spesso hanno talvolta affiancato allo stupore e all’ottimismo della scoperta, un’incertezza metodologica condivisa con altri campi del Sapere. È innegabile quanto l’avvento ormai cinquantennale delle tecnologie informatiche a supporto della comunicazione e della rappresentazione – insieme ai cambiamenti socio-economici avvenuti – abbia posto le basi per un irreversibile salto ontologico nei processi di elaborazione della forma, eppure ancora molto spesso si riscontrano riferimenti storici che possono in qualche modo ispirare teorie e prassi attuali. In questo senso alcune letture ritenute classiche, da Valéry a Benjamin, da Arnheim a Waburg forniscono non pochi spunti sostanziali per comprendere la contemporaneità. A tal proposito i noti schizzi pedagogici di Paul Klee (1925), come tutto il compendio degli appunti che l’artista svizzero ha redatto e catalogato fin dal primo giorno di insegnamento al Bauhaus (1921-31), contengono idee e prassi declinabili nel mondo dell’informatica grafica, dei software di modellazione parametrica, ma anche in ambiti dove l’obiettivo non è il controllo della forma in quanto tale, a vantaggio del controllo del processo creativo. E un rinnovato interesse al lavoro di Klee pedagogo lo si deve anche all’accessibilità open source garantita dal Zentrum Paul Klee di Brema, che consente uno studio diretto delle fonti. Un altro spunto straordinariamente interessante è la produzione crossmediale di El Lisintskij, che con i suoi Proun anticipa – o ispira – il lavoro di alcuni dei massimi architetti del secondo Novecento. Il lavoro di El Lisintskij si connota anche per la libera tracimazione tra le arti visuali – grafica, pittura, scultura e architettura – di modalità compositive, produzione di stilemi e innovazioni figurative, ponendo le basi per una visione dell’arte che potremmo definire “totale”. Altri autori dei primi due decenni del Novecento hanno esplorato le potenzialità delle rinnovate arti figurative: dal fortunato Punto, linea, superficie. di Kandinskij alla Teoria della forma e della composizione di Itten, la pedagogia delle arti del Bauhaus è, evidentemente, ancora valida come substrato culturale per la ricerca in campo architettonico, ma forse più in generale, nelle arti figurative. Ma se l’assetto pedagogico del Bauhaus e gli sviluppi del costruttivismo russo sono ben rintracciabili in letteratura, proprio dallo scambio con queste due grandi esperienze avanguardistiche, tutta una produzione grafico-artistica-performativa in area est-europea sembra abbia avuto modo di sviluppare e ibridare temi compositivi permutati da Bauhaus e Costruttivismo russo.
2018
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/374837
info:eu-repo/semantics/altIdentifier/isbn/978-84-16724-95-6
eng
ispartofbook:Sobre la influencia de las herramientas de representación e ideación en la arquitectura
XVII Congreso Internacional EGA Expresión Gráfica Arquitectónica 2018
firstpage:1415
lastpage:1422
numberofpages:8
info:eu-repo/semantics/closedAccess
STAMPA
alleditors:Marcos Alba, Carlos Luis; Juan Gutiérrez, Pablo Jeremías; Domingo Gresa, Jorge; Oliva Meyer, Justo
ByPrint Percom S.L.
country:ESP
place:Alicante
oai:iris.unitn.it:11572/3823322023-08-20T06:59:38Zcom_11572_101873com_11572_101871col_11572_101864
A Novel Method for Determination of Suspended Particulate Matter in the Atmosphere Using Array Antennas
Salas-Sanchez, Aaron Angel
Lopez-Martin, Maria Elena
Rodriguez-Gonzalez, Juan A.
Ares-Pena, Francisco J.
Salas-Sanchez, Aaron Angel
Lopez-Martin, Maria Elena
Rodriguez-Gonzalez, Juan A.
Ares-Pena, Francisco J.
In many places, atmospheric pollution is a serious public health problem. The respiratory systems of children and the aged are particularly affected; for the most vulnerable, atmospheric pollution can be mortal. Other environmental effects include reduction of the ozone layer (which in turn favours skin cancers), and damage to the fabric of buildings and monuments. The sources of atmospheric pollution include both mobile emitters (mainly motor vehicles, but also livestock) and stationary sources (notably cement works and power stations). The main components of atmospheric pollution are (depending on source) carbon monoxide, carbon dioxide, chlorofluorocarbons (CFCs), heavy metals (mainly lead), ozone, mononitrogen oxides (NO x ), sulphur dioxide, volatile organic compounds (VOCs) and other hazardous air pollutants (HAPs) (generally associated with specific industrial activity), and airborne particulate matter. Each of these components poses a health threat.
2015
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/382332
10.1109/URSI-AT-RASC.2015.7302877
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-84959497097
info:eu-repo/semantics/altIdentifier/isbn/978-9-0900-8628-6
info:eu-repo/semantics/altIdentifier/wos/WOS:000380563800063
eng
ispartofbook:2015 1st URSI Atlantic Radio Science Conference (URSI AT-RASC)
2015 1st URSI Atlantic Radio Science Conference (URSI AT-RASC)
firstpage:1
lastpage:1
numberofpages:1
info:eu-repo/semantics/closedAccess
STAMPA
IEEE Institute of Electrical and Electronics Engineers
place:Piscataway, NJ
oai:iris.unitn.it:11572/3544242024-03-22T00:52:40Zcom_11572_101873com_11572_101871col_11572_101864
Tagging large-radius b-jets from Higgs decays dropping unneeded information
Di Luca A.
Mascione D.
Follega F. M.
Cristoforetti M.
Iuppa R.
Di Luca, A.
Mascione, D.
Follega, F. M.
Cristoforetti, M.
Iuppa, R.
Multivariate approaches used in physics analyses by the High Energy Physics community often combine high-level observables estimated by very complex algorithms. The process to select these variables is usually based on a “brute force” approach, where all available event features are tested for multiple combinations of the algorithm hyperparameters. In this work, we propose an original method based on the use of a CancelOut layer to select to give as input to a Fully Connected Neural Network. Promising results are obtained in the development of a DNN classifier to select proton-proton collisions where a boosted Higgs boson decay to two b-quarks.
2022
info:eu-repo/semantics/conferenceObject
http://hdl.handle.net/11572/354424
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85131678138
eng
ispartofbook:22nd Particles and Nuclei International Conference, PANIC 2021
PANIC 2021
volume:380
journal:POS PROCEEDINGS OF SCIENCE
Sissa Medialab Srl
place:Trieste
oai:iris.unitn.it:11572/3732282024-03-22T00:49:34Zcom_11572_101873com_11572_101871col_11572_101864
Sharing Device Identification on Images from Social Media Platforms
Tomasoni A.
Verde S.
Boato G.
Tomasoni, A.
Verde, S.
Boato, G.
social media forensic
device identification
benchmarking dataset
With social networks reaching unprecedented numbers of active users and data traffic worldwide, forensic scientists have been working to secure the trustworthiness of online information against the threat of misinformation. Several works have already investigated the possibility to trace malicious contents back to its perpetrators by following the chain of sharing operations. In real-world scenarios, however, users also have the chance to upload images in multiple ways, such as via smartphone apps or desktop browsers. Being able to detect different sharing modalities may represent a valuable and still unexplored insight for forensic purposes. Following this line, in this work we present SHADE, a first collection of real-world images shared from different devices, operating systems, and user interfaces. This database provides the forensic community with a new asset for investigating the peculiar artifacts introduced by different sharing modes, and for validating detection algorithms. The dataset was evaluated by applying a set of feature descriptors borrowed from platform provenance analysis, allowing us to reach promising results in the classification of sharing modalities.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/373228
10.1109/MMSP55362.2022.9948824
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85143641814
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-7189-3
https://ieeexplore.ieee.org/document/9948824/authors#authors
info:eu-repo/semantics/altIdentifier/wos/WOS:000893205800022
eng
ispartofbook:24th International Workshop on Multimedia Signal Processing (MMSP)
MMSP 2022
firstpage:1
lastpage:6
numberofpages:6
serie:IEEE INTERNATIONAL WORKSHOP ON MULTIMEDIA SIGNAL PROCESSING
info:eu-repo/semantics/closedAccess
IEEE
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3748152024-02-10T04:57:06Zcom_11572_101873com_11572_101871col_11572_101864
Digital Experience for the Enhancement of Cultural Heritage. VR and AR Models of the Valentin im Viertel Farmhouse
Luigini, Alessandro
Brusaporci, Stefano
Basso, Alessandro
Vattano, Starlight
Maiezza, Pamela
Trizio, Ilaria
Tata, Alessandra
Empler, T.
Fusinetti, A.
Luigini, Alessandro
Brusaporci, Stefano
Basso, Alessandro
Vattano, Starlight
Maiezza, Pamela
Trizio, Ilaria
Tata, Alessandra
Modellazione digitale, realtà virtuale, realtà aumentata, museo virtuale, patrimonio architettonico = Digital modeling, virtual reality, augmented reality, virtual museum, architectural heritage
This study deals with development of the farmhouse Valentim im Viertel digital model used as a tool for the knowledge of the South Tyrol traditional architecture. The methodological approach shows the acquisition of data and the use of VR and AR models for the enhancement of the cultural heritage.
Il saggio affronta lo studio del maso Valentim im Viertel attraverso la realizzazione del modello digitale utilizzato come strumento per la conoscenza dell’architettura della tradizione del Sud Tirolo. La metodologia adottata si basa sull’acquisizione dei dati rielaborati attraverso modelli VR e AR per la valorizzazione del patrimonio.
2019
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/374815
info:eu-repo/semantics/altIdentifier/isbn/978-88-496-1942-3
ita
ispartofbook:3D Modeling & Bim. Modelli e soluzioni per la digitalizzazione
3D MODELING & BIM Modelli e soluzioni per la digitalizzazione
firstpage:440
lastpage:466
numberofpages:27
info:eu-repo/semantics/openAccess
ELETTRONICO
alleditors:Empler, T.; Fusinetti, A.
DEI srl Tipografia del Genio Civile
country:ITA
place:Roma
oai:iris.unitn.it:11572/3642512024-03-10T11:15:30Zcom_11572_101873com_11572_101871col_11572_101864
WiP: Metamodel for Continuous Authorisation and Usage Control
Hariri, Ali
Ibrahim, Amjad
Dimitrakos, Theo
Crispo, Bruno
Hariri, Ali
Ibrahim, Amjad
Dimitrakos, Theo
Crispo, Bruno
Access control has been traditionally used to protect data and privacy. Traditional access control models (e.g., ABAC, RBAC) cannot meet modern security requirements as technologies spread over heterogeneous and dynamic environments that need continuous monitoring. Modern models such as Usage Control (UCON) introduced the concept of continuous authorisation that has a lifecycle consisting of a series of phases through which the authorisation passes during its lifetime. However, such models assume a fixed lifecycle for all authorisations, so they cannot satisfy emerging technologies (e.g., smart vehicles, zero-trust, data flow), which require various and fine-grained lifecycles. Researchers have extended existing models to meet such requirements, but all solutions remain restrictive, as they are specially tailored for specific use-cases. In this paper, we propose an extensible model for continuous authorisations and usage control. The model enables its users to customise and dynamically configure the authorisation lifecycle as required by the use-case. This adds a layer of abstraction, forming a metamodel that can be instantiated into different flavours of continuous authorisation models, each addressing specific requirements. We also show that the authorisation lifecycle can be modelled as Deterministic Finite Automaton (DFA) and expressed in a structured language used by an evaluation engine to dynamically enact and manage the lifecycle. We layout the building blocks of the proposed metamodel and devise future research directions.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/364251
10.1145/3532105.3535039
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85132382281
info:eu-repo/semantics/altIdentifier/isbn/9781450393577
https://dl.acm.org/doi/10.1145/3532105.3535039
eng
ispartofbook:WiP: Metamodel for Continuous Authorisation and Usage Control
ACM SACMAT
firstpage:43
lastpage:48
numberofpages:6
info:eu-repo/semantics/closedAccess
Association for Computing Machinery (ACM)
place:New York
oai:iris.unitn.it:11572/3605062024-03-23T17:36:22Zcom_11572_101873com_11572_101871col_11572_101864
PIEMAP: Personalized Inverse Eikonal Model from Cardiac Electro-Anatomical Maps
Grandits T.
Pezzuto S.
Lubrecht J. M.
Pock T.
Plank G.
Krause R.
Grandits, T.
Pezzuto, S.
Lubrecht, J. M.
Pock, T.
Plank, G.
Krause, R.
Electroanatomical mapping, a keystone diagnostic tool in cardiac electrophysiology studies, can provide high-density maps of the local electric properties of the tissue. It is therefore tempting to use such data to better individualize current patient-specific models of the heart through a data assimilation procedure and to extract potentially insightful information such as conduction properties. Parameter identification for state-of-the-art cardiac models is however a challenging task. In this work, we introduce a novel inverse problem for inferring the anisotropic structure of the conductivity tensor, that is fiber orientation and conduction velocity along and across fibers, of an eikonal model for cardiac activation. The proposed method, named PIEMAP, performed robustly with synthetic data and showed promising results with clinical data. These results suggest that PIEMAP could be a useful supplement in future clinical workflowss of personalized therapies.
2021
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/360506
10.1007/978-3-030-68107-4_8
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85101529558
eng
ispartofbook:Statistical Atlases and Computational Models of the Heart. M&Ms and EMIDEC Challenges: 11th International Workshop, STACOM 2020, Held in Conjunction with MICCAI 2020, Lima, Peru, October 4, 2020, Revised Selected Papers
STACOM
volume:12592
firstpage:76
lastpage:86
numberofpages:11
serie:LECTURE NOTES IN COMPUTER SCIENCE
STAMPA
Springer
place:Cham
oai:iris.unitn.it:11572/3814692024-03-05T01:15:31Zcom_11572_101873com_11572_101871col_11572_101864
Emerging models for landscape digital representation: comparing experiences towards the Digital Twin of an Italian inner alpine valley
Chioni, Chiara
Favargiotti, Sara
Fistola, R.
Fregolent, L.
Rossetti, S.
La Greca, P.
Chioni, Chiara
Favargiotti, Sara
fragile territories, information technology, digitalization
Nowadays, the urban design, planning and environmental management of areas suffering territorial imbalances (namely the “peripheral”, “marginal” and “inner” areas) require exploiting the ever-increasing availability of real-time data about the natural and built environments, while promoting stakeholder engagement and empowering local communities, towards a network of truly “smart” systems. In this framework, the contribution briefly presents ongoing research projects and emerging practices that develop multidimensional models of urban and landscape systems, combining technological innovations and embedding socio-ecological perspectives. The goal of this overview is to explore the transferability of the idea of a territorial Digital Twin to the case of the Italian inner areas, conceptualising a dynamic and responsive model of the landscape, fed by an updated stream of open data, to support policy decision-making and participative planning. The case study for this experimentation is the Val di Sole, an Italian inner mountain valley in the Trentino-Alto Adige region, where to enhance the social-spatial resilience and the capacity to respond more sustainably to hazards.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/381469
info:eu-repo/semantics/altIdentifier/isbn/978-88-99237-43-1
http://www.planum.bedita.net/planum-magazine/planum-publisher-publication/atti-della-xxiv-conferenza-nazionale-siu-volume-uno-la-greca-rossetti-fistola-fregolent
eng
ispartofbook:Innovazioni tecnologiche e qualità urbana: atti della XXIV Conferenza Nazionale SIU: Dare valore ai valori in urbanistica, vol. 1
XXIV Conferenza Nazionale SIU
firstpage:153
lastpage:159
numberofpages:7
info:eu-repo/semantics/openAccess
alleditors:Fistola, R.; Fregolent, L.; Rossetti, S.; La Greca, P.
Planum Publisher; Società Italiana degli Urbanisti
country:ITA
place:Roma; Milano
oai:iris.unitn.it:11572/3703492023-12-20T00:21:05Zcom_11572_101873com_11572_101871col_11572_101864
Enabling Cooperative Autonomous Driving Through mmWave and Reconfigurable Intelligent Surfaces
Segata, Michele
Lestas, Marios
Casari, Paolo
Saeed, Taqwa
Tyrovolas, Dimitrios
Karagiannidis, George K.
Liaskos, Christos
Segata, Michele
Lestas, Mario
Casari, Paolo
Saeed, Taqwa
Tyrovolas, Dimitrio
Karagiannidis, George K.
Liaskos, Christos
Future cooperative autonomous vehicles will be able to organize into flexible platoons to improve both the efficiency and the safety of driving. However, platooning requires dependable coordination through the periodic wireless exchange of control messages. Therefore, challenging propagation scenarios as found, e.g., in dense urban areas, may hinder coordination and lead to undesirable vehicle behavior. While reconfigurable intelligent surfaces (RISs) have been advocated as a solution to improper coverage issues, no system-level simulation exists that accounts for realistic road mobility and communication aspects. To this end, we present one such simulator built on top of the OMNeT++-based PLEXE and Veins frameworks. Specifically, our contribution is a simulator that takes into account vehicle mobility, physical layer propagation, RIS coding, and networking protocols. To test our simulator, we implement an RIS-assisted autonomous platoon merging maneuver taking place at an intersection where the absence of any RIS would limit successful communications to an area dangerously close to the intersection itself. Our results validate the simulator as a feasible tool for system-level RIS-assisted cooperative autonomous vehicle maneuvering, and ultimately show the benefit of RIS as roadside infrastructure for wireless coverage extension.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/370349
10.23919/WONS57325.2023.10062109
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85150683797
info:eu-repo/semantics/altIdentifier/isbn/978-3-903176-56-0
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-2026-8
https://ieeexplore-ieee-org.ezp.biblio.unitn.it/document/10062109
info:eu-repo/semantics/altIdentifier/wos/WOS:000981938200006
eng
ispartofbook:18th IEEE/IFIP Conference on Wireless On demand Network Systems and Services (WONS 2023)
Wireless On demand Network Systems and Services Conference
firstpage:32
lastpage:39
numberofpages:8
info:eu-repo/semantics/openAccess
Institute of Electrical and Electronics Engineers (IEEE)
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3651912024-03-05T02:05:32Zcom_11572_101873com_11572_101871col_11572_101864
Probability Distributions as a Litmus Test to Inspect NNs Grounding Skills
Lucassen, A. J.
Testoni, A.
Bernardi, Raffaella
Lucassen, A. J.
Testoni, A.
Bernardi, Raffaella
Referential Guessing Games, Soft-labels, Interpretable and Trustworthy Agents.
Today AI systems are trained by ultimately using a classifier to perform a down-streaming task and are
mostly evaluated on the task-success they reach. Not enough attention is given to how the classifier
distributes the probabilities among the candidates out of which the target with the highest probability is
selected. We propose to take the probability distribution as a litmus test to inspect models’ grounding
skills. We take a visually grounded referential guessing game as test-bed and use the probability
distribution as a way to evaluate whether question answer pairs are well grounded by the model. To
this end, we propose a method to obtain such soft-labels automatically and show they correlate well
with human uncertainty about the grounded interpretation of the QA pair. Our result shows that higher
task accuracy does not necessarily correspond to a more meaningful probability distribution; we do not
consider trustworthy the models which do not pass our litmus test.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/365191
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85143255573
https://ceur-ws.org/Vol-3287/
eng
ispartofbook:Sixth Workshop on Natural Language for Artificial Intelligence
NL4AI
volume:3287
firstpage:96
lastpage:114
numberofpages:19
serie:CEUR WORKSHOP PROCEEDINGS
info:eu-repo/semantics/openAccess
ELETTRONICO
RWTH Aachen
country:DEU
place:Aachen
oai:iris.unitn.it:11572/3644222023-04-03T08:12:35Zcom_11572_101873com_11572_101871col_11572_101864
GeoFedBench: A benchmark for federated GeoSPARQL query processors
Troumpoukis, Antonis
Konstantopoulos, Stasinos
Mouchakis, Giannis
Prokopaki-Kostopoulou, Nefeli
Paris, Claudia
Bruzzone, Lorenzo
Pantazi, Despina-Athanasia
Koubarakis, Manolis
Taylor, K.
Gonçalves, R.
Lecue, F.
Yan, J.
Troumpoukis, Antoni
Konstantopoulos, Stasino
Mouchakis, Gianni
Prokopaki-Kostopoulou, Nefeli
Paris, Claudia
Bruzzone, Lorenzo
Pantazi, Despina-Athanasia
Koubarakis, Manolis
Benchmarking
Federated querying
GeoSPARQL
Performance benchmarks are invaluable for evaluating and comparing federated query processing systems, but it is hard to design benchmarks that are both realistic and informative about the systems being tested. In this paper we present GeoFedBench, a benchmark that has been obtained from an actual, practical application of geospatial and linked data querying and uses GeoSPARQL constructs that challenge all phases of federated query processing. The benchmark is publicly available as part of the Kobe suite.
2020
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/364422
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85096227825
https://ceur-ws.org/Vol-2721/
eng
ispartofbook:Proceedings of the ISWC 2020 Demos and Industry Tracks: From Novel Ideas to Industrial Practice co-located with 19th International Semantic Web Conference
19th International Semantic Web Conference on Demos and Industry Tracks: From Novel Ideas to Industrial Practice, ISWC-Posters 2020
volume:2721
firstpage:229
lastpage:232
numberofpages:4
serie:CEUR WORKSHOP PROCEEDINGS
info:eu-repo/grantAgreement/EC/H2020/825258
info:eu-repo/semantics/openAccess
alleditors:Taylor, K.; Gonçalves, R.; Lecue, F.; Yan, J.
CEUR-WS
place:Aachen
oai:iris.unitn.it:11572/3558832023-04-20T01:05:27Zcom_11572_101873com_11572_101871col_11572_101864
Performance Evaluation of Contention-based Channel Access for mmWave Sidelink Communications
Alessandro Brighenti
Matteo Drago
Tommaso Zugno
Michele Zorzi
Paolo Casari
Brighenti, Alessandro
Drago, Matteo
Zugno, Tommaso
Zorzi, Michele
Casari, Paolo
One of the main challenges of future automotive networks is the need to make vehicles aware of their surroundings. Each car will be required to collect data about the environment through dedicated sensors, and share it with its neighbors. Communicating in the millimeter wave spectrum could provide a solution for addressing such requirements. The huge amount of bandwidth available at millimeter wave frequencies, along with an optimized use of the physical resources, could provide massive data rates and low latency capabilities and enable the dissemination of real-time information. In this paper, we focus on platoons of vehicles that share LiDAR point-clouds with their platoon leader, and we use MilliCar, the ns-3 module based on the 3GPP NR V2X specifications, to provide an end-to-end performance evaluation. In particular, we study the trade-offs between using a semi-persistent resource allocation of time slots, with respect to a contention-based approach. By comparing different scheduling alternatives and different clear channel assessment thresholds, we show that coordination among different platoons can mitigate the inter-platoon interference and increase the reliability, whereas a contention-based approach achieves lower transmission delay.
2022
info:eu-repo/semantics/conferenceObject
http://hdl.handle.net/11572/355883
eng
ispartofbook:Proc. IEEE CAMAD
CAMAD
firstpage:1
lastpage:6
numberofpages:6
info:eu-repo/grantAgreement/EC/H2020/861222
info:eu-repo/semantics/openAccess
IEEE
place:United States
oai:iris.unitn.it:11572/4032922024-03-22T00:59:07Zcom_11572_101873com_11572_101871col_11572_101864
Combining a mobile deep neural network and a recurrent layer for violence detection in videos
Contardo, P.
Tomassini, S.
Falcionelli, N.
Dragoni, A. F.
Sernani, P.
Contardo, P.
Tomassini, S.
Falcionelli, N.
Dragoni, A. F.
Sernani, P.
Action Recognition
Convolutional Neural Network
Crime Investigation
Deep Learning
Law Enforcement
Long Short-Term Memory
MobileNetV2
Violence Detection
Several techniques for the automatic detection of violent scenes in videos and security footage appeared in recent years, for example with the goal of unburdening authorities from the need of analyzing hours of Closed-Circuit TeleVision (CCTV) clips. In this regard, Deep Learning-based techniques such as Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs) emerged as effective for violence detection. Nevertheless, most of such techniques require significant computational and memory resources to run the automatic detection of violence. Thus, we propose the combination of an established CNN, MobileNetV2, designed for the use in mobile and embedded devices with a recurrent layer to extract the spatio-temporal features in the security videos. A lightweight model can run in embedded devices, in a edge computing fashion, for example to allow processing the videos near the camera recording them, to preserve privacy. Specifically, we exploit transfer learning, as we use a pre-trained version of MobileNetV2, and we propose two different models combining it with a Bidirectional Long Short-Term Memory (Bi-LSTM) and a Convolutional LSTM (ConvLSTM). The paper presents accuracy tests of the two models on the AIRTLab dataset and a comparison with more complex models developed in our previous work, in order to evaluate the drop of accuracy necessary to use a model compatible with limited resources. The network composed of MobileNetV2 and the ConvLSTM scores a 94.1% accuracy, against the 96.1% of a model based on a more complex 3D CNN.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/403292
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85161927368
https://ceur-ws.org/Vol-3402/
eng
ispartofbook:CEUR Workshop Proceedings
5th International Conference on Recent Trends and Applications in Computer Science and Information Technology, RTA-CSIT 2023
volume:3402
firstpage:35
lastpage:43
numberofpages:9
serie:CEUR WORKSHOP PROCEEDINGS
info:eu-repo/semantics/openAccess
ELETTRONICO
CEUR-WS
country:DEU
place:Aachen, Germany
oai:iris.unitn.it:11572/3699472024-03-06T04:08:19Zcom_11572_101873com_11572_101871col_11572_101864
Cold-Formed Perforated Uprights: Experimental Evaluation of M-N Domains
Baldassino, N.
Bernardi, M.
Zandonini, R.
Giżejowski, M.A., Kozłowski, A.
Chybiński, M.
Rzeszut, K.
Studziński, R.
Szumigała, M.
Baldassino, N.
Bernardi, M.
Zandonini, R.
Adjustable pallet rack systems are framed structures made of cold-formed steel profiles. The columns of these systems usually are open mono-symmetric perforated profiles. Their design ask for a mixed approach, which combines experimental and analytical evaluations. In service conditions, uprights are mainly subjected to a combination of axial force and bending moments. The main features of these profiles and their variability in terms of sections geometry and layout of perforations stress the need of ad hoc investigations. The study presented in this paper focuses on a typical commercial rack upright, which was ex-perimentally investigated under ‘pure compression’ and under a combination of axial load and bending moment, assigned by applying eccentric compression loads. Specimens with a length of 1 meter were studied and a total of 9 eccentricities were considered. Results al-lowed drawing the normalized M-N domains, pointing out the remarkable interaction between axial load and bending moment.
2021
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/369947
10.1201/9781003132134-17
info:eu-repo/semantics/altIdentifier/isbn/9781003132134
info:eu-repo/semantics/altIdentifier/isbn/9780367676377
eng
ispartofbook:Modern Trends in Research on Steel, Aluminium and Composite Structures - Proceedings of the XIV International Conference on Metal Structures
ICMS2021
firstpage:157
lastpage:163
numberofpages:7
info:eu-repo/semantics/closedAccess
ELETTRONICO
alleditors:Giżejowski, M.A., Kozłowski, A.; Chybiński, M.; Rzeszut, K.; Studziński, R.; Szumigała, M.
Routledge
place:Leiden
oai:iris.unitn.it:11572/3613052024-03-24T07:55:48Zcom_11572_101873com_11572_101871col_11572_101864
Unsupervised Domain Adaptation for Video Transformers in Action Recognition
da Costa, Victor G. Turrisi
Zara, Giacomo
Rota, Paolo
Oliveira-Santos, Thiago
Sebe, Nicu
Murino, Vittorio
Ricci, Elisa
da Costa, Victor G. Turrisi
Zara, Giacomo
Rota, Paolo
Oliveira-Santos, Thiago
Sebe, Nicu
Murino, Vittorio
Ricci, Elisa
Adaptation models, Visualization, Computer vision, Source coding, Computer architecture, Benchmark testing, Transformers
Over the last few years, Unsupervised Domain Adaptation (UDA) techniques have acquired remarkable importance and popularity in computer vision. However, when compared to the extensive literature available for images, the field of videos is still relatively unexplored. On the other hand, the performance of a model in action recognition is heavily affected by domain shift. In this paper, we propose a simple and novel UDA approach for video action recognition. Our approach leverages recent advances on spatio-temporal transformers to build a robust source model that better generalises to the target domain. Furthermore, our architecture learns domain invariant features thanks to the introduction of a novel alignment loss term derived from the Information Bottleneck principle. We report results on two video action recognition benchmarks for UDA, showing state-of-the-art performance on HMDB ↔ UCF, as well as on Kinetics→NEC-Drone, which is more challenging. This demonstrates the effectiveness of our method in handling different levels of domain shift. The source code is available at https://github.com/vturrisi/UDAVT.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/361305
10.1109/ICPR56361.2022.9956679
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85143640146
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-9062-7
https://ieeexplore.ieee.org/document/9956679/authors#authors
info:eu-repo/semantics/altIdentifier/wos/WOS:000897707601037
eng
ispartofbook:International Conference on Pattern Recognition
26th International Conference on Pattern Recognition (ICPR)
firstpage:1258
lastpage:1265
numberofpages:8
serie:PROCEEDINGS INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION
info:eu-repo/semantics/closedAccess
IEEE
place:New York
oai:iris.unitn.it:11572/3551212023-04-03T08:00:55Zcom_11572_101873com_11572_101871col_11572_101864
Migration, gender and the distribution of paid and unpaid labour: preliminary perspectives on foreign couples in Italy
Molinari, Rocco
Vitali, Agnese
Gallo, Ester
Balzanella, A.
Bini, M.
Cavicchia, C.
Verde, R.
Molinari, Rocco
Vitali, Agnese
Gallo, Ester
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/355121
info:eu-repo/semantics/altIdentifier/isbn/9788891932310
https://it.pearson.com/docenti/universita/partnership/sis.html
eng
ispartofbook:SIS 2022: Book of the Short Papers
51st Scientific Meeting of the Italian Statistical Society
firstpage:687
lastpage:694
numberofpages:8
info:eu-repo/semantics/closedAccess
ELETTRONICO
alleditors:Balzanella, A.; Bini, M.; Cavicchia, C.; Verde, R.
Pearson
place:s.l.
oai:iris.unitn.it:11572/4008092024-03-19T03:27:34Zcom_11572_101873com_11572_101871col_11572_101864
Measuring and Representing Indoor Soundscapes: Insights for a Methodology from Socio-Acoustic Surveys in Residential Buildings in England
Torresin, Simone
Aletta, Francesco
Oberman, Tin
Kang, Jian
Torresin, Simone
Aletta, Francesco
Oberman, Tin
Kang, Jian
indoor soundscape, post occupancy evaluation, indoor environmental quality, acoustics, residential
The soundscape framework is effective in characterising the sound environment as we perceive it. Efforts to improve the urban soundscape must be coordinated with building design to enhance the sound environment where people spend most of their time, i.e., buildings. But how to measure the indoor soundscape? And how to represent it? Building on ISO 12913 standard series, a measurement system was defined to assess the affective response to the indoor soundscape in residential buildings, indicating the perceptual constructs to be measured and the attributes to be employed in occupant surveys. The system was first applied in a monitoring campaign in residential buildings during summer 2022, involving socio-acoustic surveys in 61 dwellings in England (UK). This paper describes the methodology employed to measure the soundscape inside buildings, both through “instruments” and “people”, and to collect contextual (e.g., information about the window view) and personal (e.g., noise sensitivity) features that may influence the perception of the acoustic environment. The results of the collected affective responses and their representation in the perceptual comfort-content reference system are presented. Representation methods are illustrated with reference to recent tools developed for outdoor soundscapes and their usefulness is demonstrated in the context of indoor soundscape design.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/400809
10.61782/fa.2023.0097
info:eu-repo/semantics/altIdentifier/isbn/978-88-88942-67-4
eng
ispartofbook:Proceedings of Forum Acusticum 2023 10th Convention of EAA
ForumAcusticum 2023
firstpage:2295
lastpage:2298
numberofpages:4
info:eu-repo/semantics/closedAccess
European Acoustics Association
place:Torino
oai:iris.unitn.it:11572/3633452024-03-10T02:58:46Zcom_11572_101873com_11572_101871col_11572_101864
Accademie e scuole di musica all’Aquila tra Sette e Ottocento
ZIMEI F
Maccavino, N.
Zimei, F. ... [et al.]
CAFIERO, R
MAIONE, P
Zimei, F
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/363345
info:eu-repo/semantics/altIdentifier/isbn/978-88-89491-28-7
ita
ispartofbook:La formazione musicale nel Meridione d’Italia tra viceregno e regno
La formazione musicale nel Meridione d’Italia tra viceregno e regno (sec. XVII-XIX)
firstpage:229
lastpage:274
numberofpages:46
serie:I TURCHINI SAGGI
info:eu-repo/semantics/closedAccess
alleditors:CAFIERO, R; MAIONE, P
Turchini Edizioni
place:Napoli
oai:iris.unitn.it:11572/3526822023-12-20T00:21:17Zcom_11572_101873com_11572_101871col_11572_101864
Multimodal Across Domains Gaze Target Detection
Tonini, Francesco
Beyan, Cigdem
Ricci, Elisa
Tumuluri, R.
Sebe, N.
Pingali, G.
Jayagopi, D.B.
Dhall, A.
Singh, R.
Anthony, L.
Salah, A.A.
Tonini, Francesco
Beyan, Cigdem
Ricci, Elisa
This paper addresses the gaze target detection problem in single images captured from the third-person perspective. We present a multimodal deep architecture to infer where a person in a scene is looking. This spatial model is trained on the head images of the person-of- interest, scene and depth maps representing rich context information. Our model, unlike several prior art, do not require supervision of the gaze angles, do not rely on head orientation information and/or location of the eyes of person-of-interest. Extensive experiments demonstrate the stronger performance of our method on multiple benchmark datasets. We also investigated several variations of our method by altering joint-learning of multimodal data. Some variations outperform a few prior art as well. First time in this paper, we inspect domain adaption for gaze target detection, and we empower our multimodal network to effectively handle the domain gap across datasets. The code of the proposed method is available at https://github.com/francescotonini/multimodal-across-domains-gaze-target-detection.
2022
info:eu-repo/semantics/conferenceObject
http://hdl.handle.net/11572/352682
10.1145/3536221.3556624
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85142777024
info:eu-repo/semantics/altIdentifier/isbn/9781450393904
https://icmi.acm.org/2022/
info:eu-repo/semantics/altIdentifier/wos/WOS:001074464500047
eng
ispartofbook:The 24th ACM International Conference on Multimodal Interaction (ICMI 2022)
ACM ICMI
firstpage:420
lastpage:431
numberofpages:12
info:eu-repo/grantAgreement/EC/H2020/871245
info:eu-repo/semantics/openAccess
alleditors:Tumuluri, R.; Sebe, N.; Pingali, G.; Jayagopi, D.B.; Dhall, A.; Singh, R.; Anthony, L.; Salah, A.A.
ACM
place:New York, USA
oai:iris.unitn.it:11572/3790942024-01-26T23:47:37Zcom_11572_101873com_11572_101871col_11572_101864
Experimental Investigation of Steel Frames Equipped with Dissipative Replaceable Links
Giuliani, Giulia
Andreotti, Roberto
Tondini, Nicola
Bonelli, Alessio
Giuliani, Giulia
Andreotti, Roberto
Tondini, Nicola
Bonelli, Alessio
In the last decades, high priority has been given to community disaster resilience owing to seismic events with a particular focus on the post-disaster restoration. Therefore, damage reduction of structural and non-structural elements after a disaster is fundamental for costs and for functionality aspects. In this context, the European RFCS project DISSIPABLE was funded with the aim to perform large demonstration tests of steel frames equipped with easily repairable seismic dissipative devices. In this paper, the experimental tests performed according to dynamic
substructuring are described. The capacity of withstanding seismic actions as well as the energy dissipation relies on the Dissipative Replaceable Link Frame system, composed of two rigid columns connected by weakened beams. Two different configurations of frames equipped with DRLF systems were tested: i) frames made of only mild steel and ii) frames made of both mild and high-strength steel. Bidimensional frames were tested under different seismic intensity
levels: Damage Limitation, Significant Damage and Near Collapse.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/379094
https://onlinelibrary.wiley.com/doi/10.1002/cepa.1811
eng
ispartofbook:International Colloquium on Stability and Ductility of Steel Structures
International Colloquium on Stability and Ductility of Steel Structures
firstpage:714
lastpage:719
numberofpages:6
info:eu-repo/semantics/openAccess
Ernst & Sohn GmbH & Co. KG
place:Berlin
oai:iris.unitn.it:11572/3747692024-03-16T01:45:10Zcom_11572_101873com_11572_101871col_11572_101864
Intelligent Transportation Systems in Smart City: A Systematic Survey
Muhammad Abul Hassan
Roofia Javed
Farhatullah
Fabrizio Granelli
Xin Gen
Muhammad Rizwan
Syed Haider Ali
Hazrat Junaid
Sana Ullah
Muhammad Abul Hassan, Roofia Javed, Farhatullah, Fabrizio Granelli, Xin Gen, Muhammad Rizwan, Syed Haider Ali , Hazrat Junaid and Sana Ullah
Hassan, Muhammad Abul
Javed, Roofia
Farhatullah,
Granelli, Fabrizio
Gen, Xin
Rizwan, Muhammad
Haider Ali, Syed
Junaid, Hazrat
Ullah, Sana
Industries, Wireless sensor networks, Smart cities, Big Data, Smart transportation, Safety, Information and communication technology
The idea of Smart Cities (SCs) was developed to digitize conventional urban living areas and redevelop using digital equipment to enhance lifestyle and resident security. Smart cities cover a wide range of applications such as smart government, smart energy, smart transportation, smart health, and smart education. In this paper, we have examined the mobility-related application e.g., smart transportation. We have focused on related to smart transportation like traffic safety, passenger and driver safety, obstacle detection, accident prevention, crime prevention, and detection, and more importantly smart vehicles like UAVs which played a key role in making these gadgets easy to use in urban environments. For a brief overview, 63 academic publications are examined in the field of ITS and SC technologies. Furthermore, we have provided possible solutions in terms of security in the smart transportation system
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/374769
10.1109/ICRAI57502.2023.10089543
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85153574713
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-6472-7
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-6473-4
https://ieeexplore.ieee.org/xpl/conhome/10089530/proceeding
enm
ispartofbook:The 5th International Conference on Robotics and Automation in Industry
The 5th International Conference on Robotics and Automation in Industry
firstpage:1
lastpage:9
numberofpages:9
serie:International Conference on Robotics and Artificial Intelligence (ICRAI)
info:eu-repo/semantics/closedAccess
ELETTRONICO
IEEE Xplore
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3620252024-03-08T07:46:19Zcom_11572_101873com_11572_101871col_11572_101864
Caputo, Andrea; Kargina, Mariya
Caputo, Andrea
Kargina, Mariya
Caputo, Andrea
Kargina, Mariya
bibliometric
cultural intelligence
research
bibliometrix
This paper presents the science map of the research produced around cultural intelligence (CQ) from management scholars. To systematize knowledge about CQ, a bibliometric analysis of 513 articles retrieved from the Web of Science Core Collection database has been made via the bibliometrix package in R, covering a range of studies from 2002 until 2020. The analyses present results concerning the performance of CQ research published in scientific journals, the geography of CQ research, and the identification of the main articles and thematic areas of research. Future research directions are drawn from the state of the art of cultural intelligence research.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/362025
eng
ispartofbook:Academy of Management Proceedings
The 82nd Annual Meeting of the Academy of Management
volume:2022
issue:1
Academy of Management
country:USA
place:Briarcliff Manor, New York, USA
oai:iris.unitn.it:11572/3699982024-03-21T21:57:59Zcom_11572_101873com_11572_101871col_11572_101864
Who Are We Talking About? Handling Person Names in Speech Translation
Gaido, Marco
Negri, Matteo
Turchi, Marco
Salesky, Elizabeth
Federico, Marcello
Costa-jussà, Marta
Gaido, Marco
Negri, Matteo
Turchi, Marco
Recent work has shown that systems for speech translation (ST) – similarly to automatic speech recognition (ASR) – poorly handle person names. This shortcoming does not only lead to errors that can seriously distort the meaning of the input, but also hinders the adoption of such systems in application scenarios (like computer-assisted interpreting) where the translation of named entities, like person names, is crucial. In this paper, we first analyse the outputs of ASR/ST systems to identify the reasons of failures in person name transcription/translation. Besides the frequency in the training data, we pinpoint the nationality of the referred person as a key factor. We then mitigate the problem by creating multilingual models, and further improve our ST systems by forcing them to jointly generate transcripts and translations, prioritising the former over the latter. Overall, our solutions result in a relative improvement in token-level person name accuracy by 47.8% on average for three language pairs (en→es,fr,it).
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/369998
10.18653/v1/2022.iwslt-1.6
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85137493637
info:eu-repo/semantics/altIdentifier/isbn/978-1-955917-41-4
https://aclanthology.org/2022.iwslt-1.6/
info:eu-repo/semantics/altIdentifier/wos/WOS:000846899900006
eng
ispartofbook:Proceedings of the 19th International Conference on Spoken Language Translation (IWSLT 2022)
19th International Conference on Spoken Language Translation, IWSLT 2022
firstpage:62
lastpage:73
numberofpages:12
info:eu-repo/semantics/openAccess
alleditors:Salesky, Elizabeth; Federico, Marcello; Costa-jussà, Marta
Association for Computational Linguistics (ACL)
place:Dublin, Ireland (in-person and online)
oai:iris.unitn.it:11572/3602222024-03-23T01:13:41Zcom_11572_101873com_11572_101871col_11572_101864
From multiresolution to the System-by-Design based GPR imaging
Zardi, Francesco
Poli, Lorenzo
Massa, Andrea
Zardi, Francesco
Poli, Lorenzo
Massa, Andrea
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/360222
10.5194/egusphere-egu22-5813
https://meetingorganizer.copernicus.org/EGU22/EGU22-5813.html
eng
ispartofbook:2022 European Geoscience Union General Assembly (EGU-GA 2022)
2022 European Geoscience Union General Assembly (EGU-GA 2022)
firstpage:[1
lastpage:2]
numberofpages:2
info:eu-repo/semantics/openAccess
Copernicus GmbH
place:Göttingen
oai:iris.unitn.it:11572/3650082024-03-23T01:00:13Zcom_11572_101873com_11572_101871col_11572_101864
Language without shared meaning : preliminary considerations on the evolutionary benefits of subjectivity
Kabbach, Alexandre
Herbelot, Aurelie
Ravignani, A.
Asano, R.
Valente, D.
Ferretti, F.
Hartmann. S.
Hayashi, M.
Jadoul, Y.
Martins, M.
Oseki, Y.
Rodrigues, E. D.
Vasileva, O.
Wacewicz, S.
Kabbach, Alexandre
Herbelot, Aurelie
We propose a characterization of language that does not rest on the hypothesis that meaning is
necessarily shared across interlocutors, since it is fundamentally grounded in the privacy and
subjectivity of mental content. We rst argue that the function of language is thus best characterized as the coordination rather than the communication between minds, since it is the constraints on use—rather than on meaning itself—that are negotiated between interlocutors during linguistic interactions. We then explore the evolutionary benets of subjectivity and argue that it positively contributes to adaptability through: 1) innovation, as conceptual variability at the
individual level increases the likelihood of the group of nding relevant conceptual innovations when exposed to environmental challenges; and 2) transfer, as subjective coordination allows for individual innovations to spread across a whole population, without individuals having to align their conceptual spaces and thereby lose the benets of conceptual variability.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/365008
10.17617/2.3398549
https://archive-ouverte.unige.ch/unige:163930
eng
ispartofbook:The evolution of language: Proceedings of the Joint Conference on Language Evolution (JCoLE)
Joint conference on language evolution (JCoLE)
firstpage:371
lastpage:378
numberofpages:8
serie:Proceedings of the Joint Conference on Language Evolution (JCoLE)
info:eu-repo/semantics/openAccess
alleditors:Ravignani, A. ; Asano, R. ; Valente, D.; Ferretti, F.; Hartmann. S.; Hayashi, M. ; Jadoul, Y. ; Martins, M. ; Oseki, Y. ; Rodrigues, E. D. ; Vasileva, O.; Wacewicz, S.
Max Planck Institute for Psycholinguistics
place:Nijmegen
oai:iris.unitn.it:11572/3716362024-03-22T19:26:22Zcom_11572_101873com_11572_101871col_11572_101864
RFID-based robot localisation: an unconstrained optimisation problem by exploiting RSSI
Shamsfakhr, F
Macii, D
Fontanelli, D
Motroni, A
Nepa, P
Palopoli, L
Buffi, A
Shamsfakhr, F
Macii, D
Fontanelli, D
Motroni, A
Nepa, P
Palopoli, L
Buffi, A
RFID robot
Robot Localisation
RSSI-based Localisation
This paper presents a Linear Least-Squares approach to estimate the pose of a mobile robot based on passive Radio Frequency Identification (RFID) technology at the Ultra-High Frequency (UHF) band. The mobile robot leverages the received signal power to obtain ranging data from a set of reference landmark RFID tags. Then, an optimisation algorithm merges RFID and odometry data to reconstruct the entire robot trajectory. The proposed method does not require precise placement of the landmarks and it is therefore fast and easy to deploy. Both numerical and experimental analysis are presented.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/371636
10.1109/I2MTC48687.2022.9806604
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85134430517
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-8360-5
https://ieeexplore.ieee.org/document/9806604/authors#authors
info:eu-repo/semantics/altIdentifier/wos/WOS:000844585400135
eng
ispartofbook:Proc. IEEE Int. Instrumentation and Measurement Technology Conference (I2MTC)
I2MTC
firstpage:1
lastpage:6
numberofpages:6
serie:CONFERENCE PROCEEDINGS - IEEE INSTRUMENTATION/MEASUREMENT TECHNOLOGY CONFERENCE
info:eu-repo/semantics/closedAccess
IEEE
place:345 E 47TH ST, NEW YORK, NY 10017 USA
oai:iris.unitn.it:11572/3716212023-05-22T07:32:09Zcom_11572_101873com_11572_101871col_11572_101864
Language Models Use Monotonicity to Assess NPI Licensing
Jumelet, J.
Denic, M.
Szymanik, J.
Hupkes, D.
Steinert-Threlkeld, S.
Jumelet, J.
Denic, M.
Szymanik, J.
Hupkes, D.
Steinert-Threlkeld, S.
We investigate the semantic knowledge of language models (LMs), focusing on (1) whether
these LMs create categories of linguistic environments based on their semantic monotonicity properties, and (2) whether these categories
play a similar role in LMs as in human language understanding, using negative polarity
item licensing as a case study. We introduce
a series of experiments consisting of probing
with diagnostic classifiers (DCs), linguistic acceptability tasks, as well as a novel DC ranking method that tightly connects the probing
results to the inner workings of the LM. By
applying our experimental pipeline to LMs
trained on various filtered corpora, we are able
to gain stronger insights into the semantic generalizations that are acquired by these models.1
2021
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/371621
10.18653/v1/2021.findings-acl.439
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85117033075
https://aclanthology.org/2021.findings-acl.439/
eng
ispartofbook:Findings of the Association of Computational Linguistics
ACL
firstpage:4958
lastpage:4969
numberofpages:12
info:eu-repo/semantics/openAccess
Association of Computational Linguistics
place:USA
oai:iris.unitn.it:11572/3699992024-03-06T04:04:32Zcom_11572_101873com_11572_101871col_11572_101864
Dodging the Data Bottleneck: Automatic Subtitling with Automatically Segmented ST Corpora
Sara Papi
Alina Karakanta
Matteo Negri
Marco Turchi
Yulan, He
Heng, Ji
Sujian, Li
Yang, Liu
Chua-Hui, Chang
Papi, Sara
Karakanta, Alina
Negri, Matteo
Turchi, Marco
Speech translation for subtitling (SubST) is the task of automatically translating speech data into well-formed subtitles by inserting subtitle breaks compliant to specific displaying guidelines. Similar to speech translation (ST), model training requires parallel data comprising audio inputs paired with their textual translations. In SubST, however, the text has to be also annotated with subtitle breaks. So far, this requirement has represented a bottleneck for system development, as confirmed by the dearth of publicly available SubST corpora. To fill this gap, we propose a method to convert existing ST corpora into SubST resources without human intervention. We build a segmenter model that automatically segments texts into proper subtitles by exploiting audio and text in a multimodal fashion, achieving high segmentation quality in zero-shot conditions. Comparative experiments with SubST systems respectively trained on manual and automatic segmentations result in similar performance, showing the effectiveness of our approach.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/369999
info:eu-repo/semantics/altIdentifier/isbn/978-1-955917-64-3
https://aclanthology.org/2022.aacl-short.59/
eng
ispartofbook:Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)
AACL | IJCNLP 2022
firstpage:480
lastpage:487
numberofpages:8
info:eu-repo/semantics/openAccess
alleditors:Yulan, He; Heng, Ji; Sujian, Li; Yang, Liu; Chua-Hui, Chang
Association for Computational Linguistics
place:Stroudsburg, PA USA
oai:iris.unitn.it:11572/3785482024-03-23T00:54:48Zcom_11572_101873com_11572_101871col_11572_101864
Data-driven Reconstruction of Rough Surfaces from Acoustic Scattering
Johnson M. -D.
Krynkin A.
Cuenca J.
Alkmim M.
De Ryck L.
Li Y.
Dolcetti G.
Johnson, M. -D.
Krynkin, A.
Cuenca, J.
Alkmim, M.
De Ryck, L.
Li, Y.
Dolcetti, G.
This work investigates the use of data-driven approaches for reconstructing rough surfaces from scattered sound. The proposed methods stands as alternatives to matrix inversion, which requires a linearisation of the dependence on the surface parameters. Here, a large dataset was formed from scattered acoustic field, estimated through the Kirchhoff Approximation. Limiting this work to the reconstruction of a static surface, K-Nearest Neighbors, Random Forests and a stochastic approach are compared to recover a parameterisation of surfaces using the scattered acoustical pressure as input. The models are then validated against a laboratory experiment alongside methods highlighted in Dolcetti et. al., JSV, 2021. The models are tested at a frequency that best fits the lab uncertainties, then tested on a broad frequency range. This scheme provides relatively accurate results in comparison to the approaches tested. Estimation errors as well as robustness in the presence of noise are discussed.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/378548
10.3397/IN_2022_0920
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85147429119
https://www.ingentaconnect.com/contentone/ince/incecp/2023/00000265/00000001/art00021#expand/collapse
eng
ispartofbook:Internoise 2022: 51st International Congress and Exposition on Noise Control Engineering
Internoise 2022
firstpage:6188
lastpage:6198
numberofpages:11
info:eu-repo/semantics/closedAccess
The Institute of Noise Control Engineering of the USA, Inc.
place:Washington
oai:iris.unitn.it:11572/3544052023-09-05T23:13:14Zcom_11572_101873com_11572_101871col_11572_101864
Vibration response of tuneable structured fabrics
Rustighi, Emiliano
Gardonio, Paolo
Cignolini, Nicola
Baldini, Sonia
Malacarne, Ciro
Perini, Marco
Rustighi, Emiliano
Gardonio, Paolo
Cignolini, Nicola
Baldini, Sonia
Malacarne, Ciro
Perini, Marco
Structured fabrics are made by interwoven rigid elements that form flexible garments such as for example chain mail armours. Although the mechanical properties of these materials are usually fixed, recent publications have demonstrated that they can be varied mechanically. In particular, it has been shown that vacuum pressure applied between two layers of 3D printed chain mails wrapped in a bag causes jamming and looking of the chain mails and this can increase the bending modulus of the material. The increase in the effective bending modulus is partly due to the compressive frictional forces and partly to the geometrical interlocking of the rigid elements complex shapes. This paper investigates for the first time the vibration response of tuneable structured fabrics when subject to variable vacuum pressures. The investigation is focused on the variation of the vibration characteristics and of the dynamical properties with the vacuum pressure. Possible applications include the realization of lightweight adaptive and semi-active vibration mitigation devices.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/354405
info:eu-repo/semantics/altIdentifier/isbn/9789082893151
https://past.isma-isaac.be/isma2022/proceedings/program/
eng
ispartofbook:Proceedings of ISMA2022 including USD2022
International Conference on Noise and Vibration Engineering 2022
info:eu-repo/semantics/closedAccess
Katholieke Universiteit Leuven
place:Leuven
oai:iris.unitn.it:11572/3699892023-05-25T12:39:24Zcom_11572_101873com_11572_101871col_11572_101864
End-to-End Speech-Translation with Knowledge Distillation: FBK@IWSLT2020
Gaido, Marco
Di Gangi, Mattia A.
Negri, Matteo
Turchi, Marco
Gaido, Marco
Di Gangi, Mattia A.
Negri, Matteo
Turchi, Marco
This paper describes FBK’s participation in the IWSLT 2020 offline speech translation (ST) task. The task evaluates systems’ ability to translate English TED talks audio into German texts. The test talks are provided in two versions: one contains the data already segmented with automatic tools and the other is the raw data without any segmentation. Participants can decide whether to work on custom segmentation or not. We used the provided segmentation. Our system is an endto-end model based on an adaptation of the Transformer for speech data. Its training process is the main focus of this paper and it is based on: i) transfer learning (ASR pertaining and knowledge distillation), ii) data augmentation (SpecAugment, time stretch and synthetic data), iii) combining synthetic and real data marked as different domains, and iv) multitask learning using the CTC loss. Finally, after the training with word-level knowledge distillation is complete, our ST models are finetuned using label smoothed cross entropy. Our best model scored 29 BLEU on the MuST-C
En-De test set, which is an excellent result compared to recent papers, and 23.7 BLEU on the same data segmented with VAD, showing the need for researching solutions addressing this specific data condition.
2020
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/369989
10.18653/v1/2020.iwslt-1.8
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85097899664
info:eu-repo/semantics/altIdentifier/isbn/978-1-952148-07-1
https://aclanthology.org/2020.iwslt-1.8/
info:eu-repo/semantics/altIdentifier/wos/WOS:000563427100008
eng
ispartofbook:Proceedings of the 17th International Conference on Spoken Language Translation
IWSLT
firstpage:80
lastpage:88
numberofpages:9
info:eu-repo/semantics/openAccess
Association for Computational Linguistics
place:Stroudsburg, PA USA
oai:iris.unitn.it:11572/3729122024-03-16T00:47:41Zcom_11572_101873com_11572_101871col_11572_101864
Comparison between Simulated and in-situ Measured Speech Intelligibility in the Multilingual Context of the Free University of Bozen-Bolzano
Torresin S.
Visentin C.
Prodi N.
Cappelletti F.
Gasparella A.
Torresin, S.
Visentin, C.
Prodi, N.
Cappelletti, F.
Gasparella, A.
Classrooms acoustics can affect studentś speech intelligibility and learning performance depending on its background noise level and/or reverberation. Speech intelligibility is usually assessed in real classrooms through a subjective approach, by performing speech intelligibility tests, or through an objective approach, by evaluating speech transmission index (STI) from impulse response, speech and noise level measurements. An acoustic simulation technique makes it possible to assess acoustical conditions for speech reception in virtual environments, thus allowing for predicting intelligibility before a classroom is built or renovated. However, in order to obtain reliable results, the simulation model needs to be calibrated and validated with in-situ measurements. The aim of this work is to compare tests performed in-situ on a group of people, with tests performed on the same people by reproducing the auralized test signal through headphones, in terms of intelligibility scores (IS), response times (RT), listening efficiency values (DE) and related STI values. Simulations have been carried out using the room acoustic software Odeon version 14.01. The investigation focused on a university classroom, which is part of the Classroom Spaces Living Lab of the Free University of Bozen-Bolzano, currently equipped with devices for monitoring energy and indoor comfort conditions, as well as detailed external weather conditions. By exploiting the bilingual context in South Tyrol, Diagnostic Rhyme Tests (DRT) in the Italian language were administered to both Italian and German native speaker students, the latter with an Italian level at least equal to B2, according to the common European framework of reference for languages. In this way, speech reception performance of the two groups has been investigated and compared.
2017
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/372912
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85050341345
info:eu-repo/semantics/altIdentifier/isbn/978-88-6046-136-0
http://www.ibpsa.org/proceedings/BSA2017/9788860461360_48.pdf
eng
ispartofbook:Building Simulation Applications, BSA 2017 - 3rd IBPSA-Italy Conference
Building Simulation Applications, BSA 2017 - 3rd IBPSA-Italy Conference
volume:2017-
firstpage:375
lastpage:383
numberofpages:9
serie:BUILDING SIMULATION APPLICATIONS BSA ...
info:eu-repo/semantics/openAccess
Free University of Bozen Bolzano
place:Bozen-Bolzano
oai:iris.unitn.it:11572/3642542023-04-03T07:44:29Zcom_11572_101873com_11572_101871col_11572_101864
SIUV: A Smart Car Identity Management and Usage Control System Based on Verifiable Credentials
Hariri, Ali
Bandopadhyay, Subhajit
Rizos, Athanasios
Dimitrakos, Theo
Crispo, Bruno
Rajarajan, Muttukrishnan
Hariri, Ali
Bandopadhyay, Subhajit
Rizos, Athanasio
Dimitrakos, Theo
Crispo, Bruno
Rajarajan, Muttukrishnan
The automotive industry is witnessing an accelerated growth in digital innovations that turn modern vehicles into digital systems. This makes the security of modern vehicles a crucial concern as they have evolved into cyber-physical and safety-critical systems. Therefore, stateful identity management and continuous access control have become a paramount requirement in smart vehicles. Indeed, several Identity and Access Management (IAM) frameworks have been proposed in the automotive field, but context awareness and continuity of control remain overlooked. To address these challenges, we present SIUV: a stateful smart-car IAM that is based on Usage Control (UCON) and Verifiable Credentials (VCs). SIUV uses Attribute Based Access Control (ABAC) policies to issue privileges to subjects (i.e. drivers or applications) according to their credentials and claims. The issued privileges are then used to decide whether to grant or deny access to in-car resources. Furthermore, the system continuously monitors subject claims, resource attributes and environmental conditions (e.g. location or time). Hence, if a change occurs, the system re-evaluates policies and updates or revokes issued privileges and usage decisions accordingly. We describe the architecture of SIUV, discuss the evaluation results, and define future directions.
2021
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/364254
10.1007/978-3-030-78120-0_3
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85111359044
info:eu-repo/semantics/altIdentifier/isbn/978-3-030-78119-4
info:eu-repo/semantics/altIdentifier/isbn/978-3-030-78120-0
https://link.springer.com/chapter/10.1007/978-3-030-78120-0_3#citeas
eng
ispartofbook:SIUV: A Smart Car Identity Management and Usage Control System Based on Verifiable Credentials
IFIP Sec
volume:625
firstpage:36
lastpage:50
numberofpages:15
serie:IFIP ADVANCES IN INFORMATION AND COMMUNICATION TECHNOLOGY
info:eu-repo/semantics/openAccess
Springer, Cham
place:New York
oai:iris.unitn.it:11572/3748342024-02-10T04:57:05Zcom_11572_101873com_11572_101871col_11572_101864
Graphic Aalysis of the Project Kina by Teresa Zarnowerowna, 1926
Vattano, Starlight
Seražin, H.
Franchini, C.
Garda, E.
Vattano, Starlight
Drawing of Architecture, digital model, Teresa Zarnowerowna, polish avantgarde art, Modern Movement
Talking about the cultural and artistic renewal in the postwar Poland one links the issue on the creation of the new generation of architects, artists, sculptors, filmmakers and poets who, coming from easteurope areas enriched by new culture of Cubism, Constructivism, Suprematism became pioneers of an eclectic culture. Among main objectives there was that of broadcasting a multiplicity of groundbreaking ideas giving shape to an artistic attitude toward the image rebuilt acconrding to the new vision, which was totally expressed at the First Exhibition of New Art in Vilnius in 1923. It was a turning point for that avantguard combined with architecture, which played a relevant role, in some cases matching with the figurative and plastic art. The article deals with the graphic study of one of the most emblematic projects of this avantgarde culture, the Project Kina by Teresa Zarnowerowna, drawn in 1926.
2018
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/374834
info:eu-repo/semantics/altIdentifier/isbn/978-961-05-0106-0
eng
ispartofbook:Women's Creativity since the Modern Movement (1918-2018): Toward a New Perception and Reception
MoMoWo Siymposium 2018-Women's Creativity since the Modern Movement (1918-2018). Toward a New Perception and Reception
firstpage:1163
lastpage:1174
numberofpages:12
info:eu-repo/semantics/openAccess
STAMPA
alleditors:Seražin, H.; Franchini, C.; Garda, E.
ZRC SAZU, France Stele Institute of Art History, Založba ZRC
country:SVN
place:Ljubljana
oai:iris.unitn.it:11572/3999142024-03-10T02:51:18Zcom_11572_101873com_11572_101871col_11572_101864
Intermittent Intelligent Camera with LEO sensor-to-satellite Connectivity
Nardello, Matteo
Caronti, Luca
Brunelli, Davide
Nardello, Matteo
Caronti, Luca
Brunelli, Davide
IoT systems can operate efficiently in scenarios with limited or sporadic power availability by utilizing intermittent power sources, such as energy harvesting. Typical implementations are based on fixed-size super-capacitors as energy storage. However, this limits the possible implementations. Large capacitors take more time to charge, resulting in extended off-time following a power failure. Small capacitors charge faster but provide a shorter active time, leading to more frequent power failures. This paper presents a fully intermittent, machine learning-based, low-power smart camera for monitoring applications integrating a two-stage energy harvester. A small first stage supports data acquisition and analysis. The second bigger energy storage is activated only for data streaming to support the integrated sub-GHz Low Earth Orbit transmission radio. This dual-stage energy storage strategy ensures both system reactivity and the capacity to sustain energy-intensive data transmission. The simulation highlights how the fully intermittent pipeline (i.e., for both the neural accelerator and the ARM core) makes the system resilient to power fluctuation and increases the throughput of processed images in ultra-low light conditions by up to 13%.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/399914
10.1145/3628353.3628550
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85180124544
info:eu-repo/semantics/altIdentifier/isbn/9798400704383
https://dl.acm.org/doi/10.1145/3628353.3628550
eng
ispartofbook:Proceedings of the 11th International Workshop on Energy Harvesting and Energy-Neutral Sensing Systems, Part of SenSys’23
ENSsys '23
firstpage:79
lastpage:85
numberofpages:7
info:eu-repo/semantics/openAccess
ACM
country:USA
place:New York, NY, USA
oai:iris.unitn.it:11572/3593342023-12-05T00:14:51Zcom_11572_101873com_11572_101871col_11572_101864
Feasibility of Dual UWB Heart Rate Sensing and Communications under FCC power restrictions
Bilich, Carlos
Bilich, Carlos
This article explores the viability of an application that can perform both, biomedical sensing and communications using the same transceiver operating at Ultra Wideband (UWB) range of frequencies. Among the numerous signals that can be measured employing UWB techniques, this work focuses on the heart rate (HR) and its variability over time (HRV). The approach exploits the fact that UWB radar sensing has long been proved possible especially for breath and HR. Founded on similar principles, UWB based communications showed great potential and some devices are already hitting the market. Still, there is no single application that can perform both tasks at the same time. A simple and approximate power budget is used to show the feasibility of sensing even under actual stringent FCC regulations and a standard communication device is chosen to show the potentialities of the combined strategy.
2007
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/359334
eng
ispartofbook:European Ultra Wide Band Radio Technology Workshop
UWB Radio Technology Workshop / IET 2007 Symposium on UWB
firstpage:1
lastpage:3
numberofpages:3
info:eu-repo/semantics/openAccess
ELETTRONICO
s.n.]
place:[S.l.
oai:iris.unitn.it:11572/4005552024-03-27T00:52:28Zcom_11572_101873com_11572_101871col_11572_101864
Weighing antimatter: AEgIS Phase 2, upgrades and first data
Volponi, Marco
Volponi, Marco
AEgIS (Antimatter Experiment: Gravity, Interferometry, Spec-troscopy) is an experiment at the Antiproton Decelerator (AD) facility at CERN whose goal is to study the asymmetry between matter and antimatter and, in par-ticular, the effect of the Earth's gravitational field on antihydrogen (Hbar ). During the 2018 run, Hbar was formed, leading into considerable gain of knowledge on the processes involved. Therefore, in the last two years, during CERN LS2, multiple upgrades to the experiment have been carried out, spanning from new degraders, a new Hbar formation trap scheme, an entirely new control system, a more efficient positronium (Ps) converter, and more efficient sensors. All this work is necessary towards the goal of creating the first pulsed beam of neutral Hbar , which will enable inertial studies on antihydrogen with high degrees of precision. This contribution presents different upgrades, their validation with the antiprotons beam, and the further developments foreseen in the future.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/400555
10.1393/ncc/i2023-23106-x
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85168014835
info:eu-repo/semantics/altIdentifier/wos/WOS:001043521100013
eng
ispartofbook:Il Nuovo Cimento C
Congresso Nazionale SIF 2022
volume:46
issue:4
journal:IL NUOVO CIMENTO C
info:eu-repo/semantics/openAccess
ELETTRONICO
SOC ITALIANA FISICA
place:Bologna
oai:iris.unitn.it:11572/4007912024-02-02T01:59:02Zcom_11572_101873com_11572_101871col_11572_101864
Iterative Superquadric Recomposition of 3D Objects from Multiple Views
Alaniz, Stephan
Mancini, Massimiliano
Akata, Zeynep
Alaniz, Stephan
Mancini, Massimiliano
Akata, Zeynep
Training, Solid modeling, Computer vision, Three-dimensional displays, Codes,
Semantics, Neural networks
Humans are good at recomposing novel objects, i.e. they can identify commonalities between unknown objects from general structure to finer detail, an ability difficult to replicate by machines. We propose a framework, ISCO, to recompose an object using 3D superquadrics as semantic parts directly from 2D views without training a model that uses 3D supervision. To achieve this, we optimize the superquadric parameters that compose a specific instance of the object, comparing its rendered 3D view and 2D image silhouette. Our ISCO framework iteratively adds new superquadrics wherever the reconstruction error is high, abstracting first coarse regions and then finer details of the target object. With this simple coarse-to-fine inductive bias, ISCO provides consistent superquadrics for related object parts, despite not having any semantic supervision. Since ISCO does not train any neural network, it is also inherently robust to out-of-distribution objects. Experiments show that, compared to recent single instance superquadrics reconstruction approaches, ISCO provides consistently more accurate 3D reconstructions, even from images in the wild. Code available at https://github.com/ExplainableML/ISCO.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/400791
10.1109/ICCV51070.2023.01651
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-0718-4
https://ieeexplore-ieee-org.ezp.biblio.unitn.it/document/10376986
eng
ispartofbook:2023 IEEE/CVF International Conference on Computer Vision (ICCV)
ICCV
firstpage:17967
lastpage:17977
numberofpages:11
serie:PROCEEDINGS IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION
info:eu-repo/semantics/openAccess
IEEE Computer Society
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3783712024-03-17T23:14:30Zcom_11572_101873com_11572_101871col_11572_101864
Optimal Sizing and Placement of Droop-based Converters in DC Microgrids With ZIP Loads
Chapaloglou, Spyridon
Abdolmaleki, Babak
Tedeschi, Elisabetta
Chapaloglou, Spyridon
Abdolmaleki, Babak
Tedeschi, Elisabetta
Current-sharing
dc microgrid
droop control
MILP optimization
voltage control
Achieving satisfactory voltage regulation in dc microgrids with distributed generators while using those as efficiently as possible, does not only depend on implementing appropriate control laws but also on making smart decisions in the planning stage. For the droop-based decentralized control framework, the droop design depends on the generator's location, the grid topology, and the load distribution, especially when constant power loads are considered. In this paper, a mixed-integer linear programming methodology is proposed for simultaneously deciding the locations and the size of distributed generators in dc microgrids to achieve optimal current sharing. The problem formulation includes a detailed discussion on the linearization of the various non-linear terms and the integration of the power flow equations. In addition, a set of constraints is proposed to help estimate the least upper bound of the generator's rating depending on the number of available generators. Then, the proposed formulation is used to explore the optimal placement and sizing of droop-based generators as a function of how many are available, revealing preference patterns associated with the microgrid structure. The proposed optimal decisions are also validated with time-domain simulations demonstrating the feasibility of the solution.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/378371
10.1109/SMART55236.2022.9990094
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85146288170
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-7146-6
https://ieeexplore.ieee.org/document/9990094
eng
ispartofbook:2022 2nd International Conference on Sustainable Mobility Applications, Renewables and Technology
SMART 2022
firstpage:1
lastpage:8
numberofpages:8
info:eu-repo/semantics/closedAccess
Institute of Electrical and Electronics Engineers Inc.
place:Piscataway, New Jersey, United States
oai:iris.unitn.it:11572/4010942024-03-02T00:51:11Zcom_11572_101873com_11572_101871col_11572_101864
A research infrastructure for generating and sharing diversity-aware data
Busso, Matteo
Abente, Ronald
De Götzen, Amalia
Murukannaiah, Pradeep K.
Hirzle, Teresa
Busso, Matteo
Abente, Ronald
De Götzen, Amalia
The intensive flow of personal data associated with the trend of computerizing aspects of people’s
diversity in their daily lives is associated with issues concerning not only people protection and their
trust in new technologies, but also bias in the analysis of data and problems in their management and
reuse. Faced with a complex problem, the strategies adopted, including technologies and services, often
focus on individual aspects, which are difficult to integrate into a broader framework, which can be
of effective support for researchers and developers. Therefore, we argue for the development of an
end-to-end research infrastructure (RI) that enables trustworthy diversity-aware data within a citizen
science community.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/401094
https://ceur-ws.org/Vol-3456/
eng
ispartofbook:Proceedings of the Workshops at the Second International Conference on Hybrid Human-Artificial Intelligence co-located with (HHAI 2023)
HHAI 2023
volume:3456
firstpage:220
lastpage:226
numberofpages:7
serie:CEUR WORKSHOP PROCEEDINGS
info:eu-repo/semantics/openAccess
alleditors:Murukannaiah, Pradeep K.; Hirzle, Teresa
RWTH Aachen
place:Aachen, Germany
oai:iris.unitn.it:11572/4007902024-02-07T12:17:37Zcom_11572_101873com_11572_101871col_11572_101864
PDiscoNet: Semantically consistent part discovery for fine-grained recognition
van der Klis, Robert
Alaniz, Stephan
Mancini, Massimiliano
Dantas, Cassio F.
Ienco, Dino
Akata, Zeynep
Marcos, Diego
van der Klis, Robert
Alaniz, Stephan
Mancini, Massimiliano
Dantas, Cassio F.
Ienco, Dino
Akata, Zeynep
Marcos, Diego
Location awareness, Visualization, Shape, Computational modeling,
Semantics, Modulation, Transforms
Fine-grained classification often requires recognizing specific object parts, such as beak shape and wing patterns for birds. Encouraging a fine-grained classification model to first detect such parts and then using them to infer the class could help us gauge whether the model is indeed looking at the right details better than with interpretability methods that provide a single attribution map. We propose PDiscoNet to discover object parts by using only image-level class labels along with priors encouraging the parts to be: discriminative, compact, distinct from each other, equivariant to rigid transforms, and active in at least some of the images. In addition to using the appropriate losses to encode these priors, we propose to use part-dropout, where full part feature vectors are dropped at once to prevent a single part from dominating in the classification, and part feature vector modulation, which makes the information coming from each part distinct from the perspective of the classifier. Our results on CUB, CelebA, and PartImageNet show that the proposed method provides substantially better part discovery performance than previous methods while not requiring any additional hyper-parameter tuning and without penalizing the classification performance. The code is available at https://github.com/robertdvdk/part_detection
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/400790
10.1109/ICCV51070.2023.00179
info:eu-repo/semantics/altIdentifier/isbn/979-8-3503-0718-4
https://ieeexplore.ieee.org/document/10378232/authors#authors
eng
ispartofbook:2023 IEEE/CVF International Conference on Computer Vision (ICCV)
ICCV
firstpage:1866
lastpage:1876
numberofpages:11
serie:PROCEEDINGS IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION
info:eu-repo/semantics/openAccess
IEEE Computer Society
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3629062024-03-20T20:33:23Zcom_11572_101873com_11572_101871col_11572_101864
QuestEval: Summarization Asks for Fact-based Evaluation
Scialom, Thomas
Dray, Paul-Alexis
Lamprier, Sylvain
Piwowarski, Benjamin
Staiano, Jacopo
Wang, Alex
Gallinari, Patrick
Scialom, Thoma
Dray, Paul-Alexi
Lamprier, Sylvain
Piwowarski, Benjamin
Staiano, Jacopo
Wang, Alex
Gallinari, Patrick
Summarization evaluation remains an open research problem: current metrics such as ROUGE are known to be limited and to correlate poorly with human judgments. To alleviate this issue, recent work has proposed evaluation metrics which rely on question answering models to assess whether a summary contains all the relevant information in its source document. Though promising, the proposed approaches have so far failed to correlate better than ROUGE with human judgments. In this paper, we extend previous approaches and propose a unified framework, named QuestEval. In contrast to established metrics such as ROUGE or BERTScore, QuestEval does not require any ground-truth reference. Nonetheless, QuestEval substantially improves the correlation with human judgments over four evaluation dimensions (consistency, coherence, fluency, and relevance), as shown in extensive experiments.
2021
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/362906
10.18653/v1/2021.emnlp-main.529
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85119949259
https://aclanthology.org/2021.emnlp-main.529
info:eu-repo/semantics/altIdentifier/wos/WOS:000860727000044
eng
ispartofbook:Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
EMNLP
firstpage:6594
lastpage:6604
numberofpages:11
info:eu-repo/semantics/openAccess
Association for Computational Linguistics
place:209 N. Eighth Street, Stroudsburg PA 18360, USA
oai:iris.unitn.it:11572/3602152023-04-03T07:52:30Zcom_11572_101873com_11572_101871col_11572_101864
Exploiting Wireless Localization for Decision Support in Search-And-Rescue Operations
Polo A.
Rocca P.
Salucci M.
Massa A.
Polo, A.
Rocca, P.
Salucci, M.
Massa, A.
Wireless localization, distributed sensing, decision support system (DSS), emergency management, search-and-rescue (SAR) mission
This paper introduces "SHARON", an innovative decision support system (DSS) that exploits wireless communication and localization technologies for the analysis and the planning of Search-And-Rescue (SAR) missions in mountain or rural environments.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/360215
10.1109/MOCAST54814.2022.9837566
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85136086543
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-6717-9
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-6718-6
https://ieeexplore.ieee.org/document/9837566
info:eu-repo/semantics/altIdentifier/wos/WOS:000853078900024
eng
ispartofbook:2022 11th International Conference on Modern Circuits and Systems Technologies
MOCAST 2022
firstpage:1
lastpage:4
numberofpages:4
info:eu-repo/semantics/closedAccess
Institute of Electrical and Electronics Engineers
place:Piscataway, NJ
oai:iris.unitn.it:11572/3534012023-04-03T07:44:50Zcom_11572_101873com_11572_101871col_11572_101864
Facebook Ads as a Demographic Tool to Measure the Urban-Rural Divide
Rama, Daniele
Mejova, Yelena
Tizzoni, Michele
Kalimeri, Kyriaki
Weber, Ingmar
Rama, Daniele
Mejova, Yelena
Tizzoni, Michele
Kalimeri, Kyriaki
Weber, Ingmar
digital demography
online advertising
social network
urban-rural divide
In the global move toward urbanization, making sure the people remaining in rural areas are not left behind in terms of development and policy considerations is a priority for governments worldwide. However, it is increasingly challenging to track important statistics concerning this sparse, geographically dispersed population, resulting in a lack of reliable, up-to-date data. In this study, we examine the usefulness of the Facebook Advertising platform, which offers a digital "census" of over two billions of its users, in measuring potential rural-urban inequalities. We focus on Italy, a country where about 30% of the population lives in rural areas. First, we show that the population statistics that Facebook produces suffer from instability across time and incomplete coverage of sparsely populated municipalities. To overcome such limitation, we propose an alternative methodology for estimating Facebook Ads audiences that nearly triples the coverage of the rural municipalities from 19% to 55% and makes feasible fine-grained sub-population analysis. Using official national census data, we evaluate our approach and confirm known significant urban-rural divides in terms of educational attainment and income. Extending the analysis to Facebook-specific user "interests" and behaviors, we provide further insights on the divide, for instance, finding that rural areas show a higher interest in gambling. Notably, we find that the most predictive features of income in rural areas differ from those for urban centres, suggesting researchers need to consider a broader range of attributes when examining rural wellbeing. The findings of this study illustrate the necessity of improving existing tools and methodologies to include under-represented populations in digital demographic studies - the failure to do so could result in misleading observations, conclusions, and most importantly, policies.
2020
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/353401
10.1145/3366423.3380118
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85086594597
info:eu-repo/semantics/altIdentifier/isbn/978-1-4503-7023-3
https://doi.org/10.1145/3366423.3380118
info:eu-repo/semantics/altIdentifier/wos/WOS:000626273300030
eng
ispartofbook:Proceedings of The World Wide Web Conference WWW 2020
WWW '20
firstpage:327
lastpage:338
numberofpages:12
info:eu-repo/semantics/openAccess
Association for Computing Machinery, Inc
country:USA
place:New York
oai:iris.unitn.it:11572/3905102023-10-13T22:56:01Zcom_11572_101873com_11572_101871col_11572_101864
Ca’ Venier e ponte dell’Accademia nel 1985. Tre immagini transitorie = Ca’ Venier and Ponte dell’Accademia in 1985. Three Transitional Images
Vattano, Starlight
Cannella, M.
Garozzo, A.
Morena, S.
Vattano, Starlight
Interpretazione grafica, rappresentazione, modellazione digitale, ponte dell’Accademia, Ca’ Venier dei Leoni = Graphic interpretation, Representation, Digital modeling, Ponte dell’Accademia, Ca’ Venier dei Leoni
The Third International Architecture Exhibition directed by Aldo Rossi as part of the 1985 Architecture Biennale represented a ield of design synthesis for Venice that gave space to hundreds of projects addressed to the Venetian terrotory posing general problems to be answered, as Rossi himself points out, with the composition of a different architecture, “by people who live in the most distant places”. On the track of a research initiated in collaboration with the direction of the Archivio
Progetti Iuav in 2021, which had as aim the knowledge dissemination of the graphic heritage and documents related to the projects stored in it. The article proposes the graphic reading and digital interpretation of some projects presented as part of the 1985 Exhibition, the one for the ponte dell’Accademia and the one for Ca’ Venier dei Leoni, taking into account the broader proposal of the unbuilt digital restitution in the 20th century. This allows to compare the trace of the Canal Grande
deining a mapping of unexplored architectural events of Venetian memory. In the reinterpretation of the design proposals presented, the digital restitution of documents and drawings preserved at the Archivio Progetti Iuav, provides unpublished images that reactivate the processes of interchange between memory and society, between cultural heritage and the contemporary city, between possibility and the real.
La Terza Mostra Internazionale di Architettura diretta da Aldo Rossi all’interno della Biennale di Architettura del 1985 rappresentò per Venezia un campo di sintesi progettuale che diede spazio a centinaia di progetti rivolti alla terraferma veneziana, ponendo problemi di ordine generale ai quali venne chiesto di rispondere, come sottolinea lo stesso Rossi, con la composizione di un’architettura diversa, eterogenea per provenienza geograica e culturale. Sulla traccia di una ricerca avviata in
collaborazione con la direzione dell’Archivio Progetti Iuav nel 2021, che ha avuto come obiettivo quello della divulgazione della conoscenza del patrimonio di disegni e documenti relativi ai progetti in esso conservati, l’articolo propone la rilettura graica e l’interpretazione digitale di alcuni dei progetti presentati nell’ambito della Terza Mostra Internazionale di Architettura del 1985. I due casi studio, il ponte dell’Accademia e Ca’ Venier dei Leoni, sono da collocare in una più ampia operazione di restituzione digitale del non costruito nel XX secolo, che si confronta con la traccia del Canal Grande e deinisce una mappatura di eventi architettonici inesplorati della memoria veneziana. Nella rilettura delle occasioni progettuali presentate, la restituzione digitale dei documenti e dei disegni conservati presso l’Archivio Progetti Iuav, fornisce immagini inedite che riattivano i processi di interscambio tra la memoria e la società, tra il patrimonio culturale e la città contemporanea, tra la possibilità e il reale.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/390510
10.3280/oa-1016-c397
info:eu-repo/semantics/altIdentifier/isbn/9788835155119
https://series.francoangeli.it/index.php/oa/catalog/book/1016
ita
ispartofbook:Transizioni Transitions. Attraversare Modulare Procedere Cross Modulate Develop
44° Convegno internazionale dei docenti delle discipline della rappresentazione congresso della Unione Italiana per il Disegno. Atti 2023 44th International Conference of Representation Disciplines Teachers Congress of Unione Italiana per il Disegno. Proceedings 2023
firstpage:2129
lastpage:2150
numberofpages:22
info:eu-repo/semantics/openAccess
ELETTRONICO
alleditors:Cannella, M.; Garozzo, A.; Morena, S.
FrancoAngeli
country:ITA
place:Milano
oai:iris.unitn.it:11572/3791042023-12-02T23:49:17Zcom_11572_101873com_11572_101871col_11572_101864
An improved Analytical Formula for Predicting the Temperature of Heavily Protected Steel Sections
Possidente, Luca
Tondini, Nicola
Wickström, Ulf
Liming Jiang
Paulo Vila Real
Xinyan Huang
Mhd Anwar Orabi
Jin Qiu
Tianwei Chu
Zhuojun Nan
Cheng Chen
Zhiruoyu Wang
Asif Usmani
Possidente, Luca
Tondini, Nicola
Wickström, Ulf
Fire protection
steel temperature
heat transfer
steel structure
Fire safety engineering
Fire verification might be particularly demanding for steel structures and insulation is a common option to slow down the temperature increase in the steel elements without modifying the original structural design. Simple analytical formulae, as provided for instance in EN1993-1-2 design standard, allow a quick estimate
of the temperature of insulated steelwork, without determining the thermal field inside a steel cross-section by performing in-depth experimental or numerical analyses. However, the EN1993-1-2 formulation considers heat transfer with temperature boundary conditions, rather than more realistic conditions on the heat flux, and is inaccurate for heavily insulated steel sections, in which protective solutions with high heat capacity are adopted. In this paper a new analytical formula aimed at estimating the temperature of protected steel members is proposed. Its accuracy is assessed by comparing the predictions of the proposed and the EN1993-1-2 formulations with the results of a parametric analysis consisting of 1-D models. Several steel thicknesses, insulation materials and thicknesses and an exposure to the ISO 834 heating curve are considered in the analyses. It is shown that the EN1993-1-2 can be both conservative and unconservative depending on the ratio between the insulation and the steel heat capacities mu and is not suited for heavily insulated steel sections with high values of mu. On the contrary, the proposed formulation results in being always safe and particularly suited for heavily insulated steel sections.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/379104
info:eu-repo/semantics/altIdentifier/isbn/978-962-367-869-8
eng
ispartofbook:SIF 2022 12th International Conference on Structures in Fire
12th International Conference on Structures in Fire
firstpage:905
lastpage:914
numberofpages:10
info:eu-repo/semantics/openAccess
alleditors:Liming Jiang; Paulo Vila Real; Xinyan Huang; Mhd Anwar Orabi; Jin Qiu; Tianwei Chu; Zhuojun Nan; Cheng Chen; Zhiruoyu Wang; Asif Usmani
The Hong Kong Polytechnic University
place:Hong Kong
oai:iris.unitn.it:11572/3748682024-02-11T01:11:14Zcom_11572_101873com_11572_101871col_11572_101864
L'impronta del doppio: il movimento immobile
Vattano, Starlight
Belardi, P.
Cirifaci, A.
Di Luggo, A.
Dotto, E.
Gai, F.
Maggio, F.
Quici, F.
Vattano, Starlight
Movimento, rappresentazione, impronta, disegno, percezione = Movement, representation, impression, drawing, perception
The moment in time renders the reality of the imprint the matter of rhythmic scansion, representation of form, drawing of gesture. The imprint is the form of a frozen movement. The article deals with a study on the perception of movement in the context of representation.
L’attimo temporale rende la realtà dell’impronta materia della scansione ritmica, rappresentazione della forma, disegno del gesto. L’impronta è forma di un movimento congelato. L'articolo affronta uno studio sulla percezione del movimento nell'ambito della rappresentazione.
2014
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/374868
info:eu-repo/semantics/altIdentifier/isbn/978-88-904585-8-3
ita
ispartofbook:Idee per la Rappresentazione 6 IMPRONTE atti del seminario di studi
Atti del Seminario di studi Impronte 6. Idee per la Rappresentazione
firstpage:98
lastpage:109
numberofpages:12
info:eu-repo/semantics/closedAccess
STAMPA
alleditors:Belardi, P.; Cirifaci, A.; Di Luggo, A.; Dotto, E.; Gai, F.; Maggio, F.; Quici, F.
Dipartimento di Storia, Disegno e Restauro della Sapienza Università Roma, Dipartimento di Ingegneria Civile e Ambientale Università degli Studi di Perugia, Università di Catania
country:ITA
place:Roma
oai:iris.unitn.it:11572/4015542024-03-08T00:33:01Zcom_11572_101873com_11572_101871col_11572_101864
An Analytical Model to Investigate the Effect of Diaphragms on the Elastic Behaviour of Multi-Storey Coupled-Panel Clt Shearwalls
Oliveira D.
Mikael A.
Casagrande D.
Doudak G.
Oliveira, D.
Mikael, A.
Casagrande, D.
Doudak, G.
CLT
Derivation
Diaphragm Effect
Earthquake
Lateral Design
Mid-rise
Modelling
This study develops an analytical model to describe the behaviour of multi-storey multi-panel CLT shearwalls, specifically accounting for cumulative loads between storeys and the effect of the diaphragm. The analysis considers the contribution of hold-downs, wall-to-floor connections, and panel-to-panel joints, as well as loads transferred from storeys above. The analysis has been developed for shearwalls acting as a series of coupled panels (CP) each individually rotating about a corner. Force transfer between storeys is implemented by distributing reactions through the upper floor diaphragm and a direct force from the upper storey's hold-down. Structuring the equations this way allows for a simplified formulation while including several components of the system's complex behaviour. Two-dimensional finite element modelling is used to verify the accuracy of the developed model.
2023
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/401554
10.52202/069179-0364
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85171796067
info:eu-repo/semantics/altIdentifier/isbn/978-1-7138-7329-7
info:eu-repo/semantics/altIdentifier/isbn/978-1-7138-7327-3
eng
ispartofbook:WCTE2023 - Timber for a Livable Future 13th World Conference on Timber Engineering Proceedings
13th World Conference on Timber Engineering: Timber for a Livable Future, WCTE 2023
volume:5
firstpage:2783
lastpage:2791
numberofpages:9
info:eu-repo/semantics/closedAccess
World Conference on Timber Engineering (WCTE)
place:Oslo, Norway
oai:iris.unitn.it:11572/3697582024-03-22T19:28:23Zcom_11572_101873com_11572_101871col_11572_101864
Lightweight Parsing and Slicing for Bug Identification in C
Mecenero, Luca
Paramitha, Ranindya
Pashchenko, Ivan
Massacci, Fabio
Mecenero, Luca
Paramitha, Ranindya
Pashchenko, Ivan
Massacci, Fabio
Program slicing has been used to semi- or fully-automatically help developers find errors and vulnerabilities in their programs. For example, Dashevskyi et al. (IEEE TSE 2018) introduced a lightweight slicer for Java that can be used for vulnerability analysis. However, a similar lightweight slicer for C/C++ is still missing. In this work we propose a comparison method for parsers, evaluate it on two commonly-used parsers, and develop a lightweight slicer for C/C++ using the “better” parser from our comparison. From our evaluation, the Joern parsing method (island grammar) could parse non-standard C/C++ code but its resulting structure may contain semantic errors that can affect subsequent analysis. ANTLR4 is faster in returning a result, and when manually cleared of non-standard C/C++ codes, it is more accurate than Joern. We then built our C/C++ thin slicer extension using ANTLR4, and we observed that it is promising from both precision and performance perspectives. As a future work, we plan to improve the logic behind processing pointers. In particular, we consider doing deeper pointer analysis.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/369758
10.1145/3538969.3543828
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85136958645
info:eu-repo/semantics/altIdentifier/isbn/9781450396707
https://dl.acm.org/doi/10.1145/3538969.3543828
eng
ispartofbook:ARES '22: Proceedings of the 17th International Conference on Availability, Reliability and Security (IWCSEC 2022)
ARES
firstpage:1
lastpage:10
numberofpages:10
info:eu-repo/semantics/openAccess
ELETTRONICO
Association for Computing Machinery
country:ITA
place:New York, NY, USA
oai:iris.unitn.it:11572/3518402024-03-12T19:55:57Zcom_11572_101873com_11572_101871col_11572_101864
Visible Light Synchronization for Time-Slotted Energy-Aware Transiently-Powered Communication
Torrisi, Alessandro
Doglioni, Maria
Yildirim, Kasim Sinan
Brunelli, Davide
Torrisi, Alessandro
Doglioni, Maria
Yildirim, Kasim Sinan
Brunelli, Davide
Energy-harvesting IoT devices that operate without batteries paved the way for sustainable sensing applications. These devices force applications to run intermittently since the ambient energy is sporadic, leading to frequent power failures. Unexpected power failures introduce several challenges to wireless communication since nodes are not synchronized and stop operating during data transmission. This paper presents a novel self-powered autonomous circuit design to remedy this problem. This circuit uses visible-light communication (VLC) to enable synchronization for time-slotted energy-aware transiently powered communication. Therefore, it aligns the activity phases of the batteryless sensors so that energy status communication occurs when these nodes are active simultaneously. Evaluations showed that our circuit has an ultra-low power consumption, can work with zero energy cost by relying only on the harvested energy, and supports efficient intermittent communication over intermittently powered nodes.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/351840
10.1145/3531437.3539722
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85136242468
info:eu-repo/semantics/altIdentifier/isbn/9781450393546
https://dl.acm.org/doi/10.1145/3531437.3539722
eng
ispartofbook:Proceedings of the ACM/IEEE International Symposium on Low Power Electronics and Design (ISLPED '22)
ISLPED '22
firstpage:1
lastpage:6
numberofpages:6
info:eu-repo/semantics/closedAccess
ACM Association for Computing Machinery
country:USA
place:New York City
oai:iris.unitn.it:11572/4016602024-03-16T02:07:58Zcom_11572_101873com_11572_101871col_11572_101864
Kinematic Modes of CLT Single-Storey Shearwalls with Openings in the Inelastic Range
Mestar M.
Doudak G.
Casagrande D.
Mestar, M.
Doudak, G.
Casagrande, D.
CLT walls with opening
Failure mechanism
Hold-downs layout
Inelastic analysi
Kinematic modes
The paper presents an investigation on the influence of geometrical dimensions of lintel beam, wall segment, lamella layup and the mechanical behaviour of hold-down on the inelastic behaviour of the CLT shearwall assembly. Five full-scale shearwalls were tested and used in the validation of a proposed numerical model. The verification of failure in the CLT panels is done using the RVSE model and the comparison between the analytical results and those obtained from tested was done using the Equivalent Energy Elastic Plastic method. The parametric study of the behaviour of the walls in the inelastic range demonstrated the important effect of the lintel beam, wall segment slenderness as well as the hold-down stiffness on the mechanical and global kinematic behaviour of the wall. It was observed that the kinematic modes can change when the walls are loaded beyond their elasticity limit. The failure mode and the global ductility were also found to be dependent on the hold-down configurations, particularly for walls with door openings.
2021
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/401660
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85120731260
info:eu-repo/semantics/altIdentifier/isbn/9781713841111
eng
ispartofbook:World Conference on Timber Engineering (WCTE 2021)
WCTE 2021
info:eu-repo/semantics/closedAccess
Curran Associates, Inc.
place:Santiago (Chile)
oai:iris.unitn.it:11572/3748492024-02-11T01:10:42Zcom_11572_101873com_11572_101871col_11572_101864
Air Movement Representation as a Tool for Urban and Architectural Environmental Quality
Germanà, Maria Luisa
Vattano, Starlight
Tavolante, Alessia
Burlando, M.
Canepa, M.
Magliocco, A.
Perini, K.
Repetto, M.P.
Germanà, Maria Luisa
Vattano, Starlight
Tavolante, Alessia
Environmental Design, Bioclimatic Architecture, Passive cooling, Air Movement
Air movement is a measurable physical phenomenon that produces indirectly measurable effects on people and on the environment. It is an essential prerequisite for the comfort and the hygiene of the built environment and it is highly relevant to the passive cooling strategies. The relationships between air movement and built environment have to be analysed under a multi-scale vision. They fall within the field of urban and architectural design because they can be influenced by artificial expedients, although they are always caused by natural factors. This paper discusses the need for a complete and objective representation of different kinds of air movement, to improve the urban and architectural design and to integrate the knowledge and the assessment of the built environment. The proposal for a graphic standard is partially described, aiming to contribute towards a codified and multi-scale representation of the air movements that can be considered as a tool for environmental quality, both in the maintenance or the conservation activities and in the transformation processes.
2017
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/374849
info:eu-repo/semantics/altIdentifier/isbn/978-88-97752-91-2
https://gup.unige.it/sites/gup.unige.it/files/pagine/URBAN-CEQ.pdf
ita
ispartofbook:International Conference on Urban Comfort and Environmental Quality URBAN CEQ 28-29 September 2017
URBAN-CEQ. International Conference on Urban Comfort and Environmental Quality
firstpage:76
lastpage:85
numberofpages:10
info:eu-repo/semantics/closedAccess
STAMPA
alleditors:Burlando, M.; Canepa, M.; Magliocco, A.; Perini, K.; Repetto, M.P.
GUP Genova University Press
country:ITA
place:Genova
oai:iris.unitn.it:11572/3602062024-03-20T19:18:34Zcom_11572_101873com_11572_101871col_11572_101864
Fine-grained Stop-Move Detection in UWB-based Trajectories
Hachem, F
Vecchia, D
Damiani, ML
Picco, GP
Hachem, F
Vecchia, D
Damiani, Ml
Picco, Gp
Ultra-wideband
mobility pattern
trajectory
Ultra-wideband (UWB) localization enables user tracking with high spatio-temporal resolution, whose exploitation for detecting higher-level mobility patterns is largely unexplored. We study whether i) existing detection techniques, developed for coarser-grained localization, apply also to UWB trajectories, and ii) the quantitative extent to which this enables finer-grained analyses. We focus on the well-known stop-move pattern, and offer a concrete use case of capturing visits in a real museum. We contribute a novel metric suited to the high UWB spatio-temporal resolution and use it to evaluate representative techniques. We deploy a UWB system in a 25 x15 m(2) museum area and base our analysis on 70000+ positions and 200+ ground-truth stops. These are very close in space and time, yet results confirm very accurate spatio-temporal estimation in the vast majority of cases.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/360206
10.1109/PerCom53586.2022.9762404
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85129989889
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-1643-6
https://ieeexplore.ieee.org/document/9762404/authors#authors
info:eu-repo/semantics/altIdentifier/wos/WOS:000835724600012
eng
ispartofbook:Proceedings of the 2022 IEEE International Conference on Pervasive Computing and Communications (PerCom)
PerCom
firstpage:111
lastpage:118
numberofpages:8
serie:PROCEEDINGS OF THE ... IEEE INTERNATIONAL CONFERENCE ON PERVASIVE COMPUTING AND COMMUNICATIONS
info:eu-repo/semantics/openAccess
IEEE
place:Piscataway, NJ USA
oai:iris.unitn.it:11572/3730962024-03-17T01:21:26Zcom_11572_101873com_11572_101871col_11572_101864
30 Gb/s NRZ Transmission with Lumped-Element Silicon Photonic Mach-Zehnder Modulator
Cammarata, Simone
Velha, Philippe
Palla, Fabrizio
Di Pasquale, Fabrizio
Saponara, Sergio
Faralli, Stefano
Cammarata, Simone
Velha, Philippe
Palla, Fabrizio
Di Pasquale, Fabrizio
Saponara, Sergio
Faralli, Stefano
Mach-Zehnder interferometers,
Modulation,
Bandwidth,
Electro-optic modulators,
Silicon photonics,
Foundries,
Optical signal processing
This work presents the experimental characterization of a compact non-traveling-wave Mach-Zehnder modulator fabricated in a standard foundry process. Non-return-to-zero modulation was validated till 30 Gb/s by BER measurements and 13.2 GHz-6-dBe electro-optic bandwidth was measured under single-arm driving.
2022
info:eu-repo/semantics/conferenceObject
https://hdl.handle.net/11572/373096
10.1109/IPC53466.2022.9975756
info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85145553143
info:eu-repo/semantics/altIdentifier/isbn/978-1-6654-3487-4
https://ieeexplore.ieee.org/document/9975756/authors#authors
eng
ispartofbook:2022 IEEE Photonics Conference (IPC)
IPC
firstpage:1
lastpage:2
numberofpages:2
serie:IEEE PHOTONICS CONFERENCE
info:eu-repo/semantics/closedAccess
IEEE
place:Piscataway, NJ USA
oai_dc///col_11572_101864/100