Understanding human behaviors in crowded scenarios requires analyzing not only the position of the subjects in space, but also the scene context. Existing approaches mostly rely on the motion history of each pedestrian and model the interactions among people by considering the entire surrounding neighborhood. In our approach, we address the problem of motion prediction by applying coherent group clustering and a global attention mechanism on the LSTM-based Generative Adversarial Networks (GANs). The proposed model consists of an attentive group-aware GAN that observes the agents' past motion and predicts future paths, using (i) a group pooling module to model neighborhood interaction, and (ii) an attention module to specifically focus on hidden states. The experimental results demonstrate that our proposal outperforms state-of-the-art models on common benchmark datasets, and is able to generate socially-acceptable trajectories.

AG-GAN: An Attentive Group-Aware GAN for pedestrian trajectory prediction / Song, Yue; Bisagno, Niccolò; Hassan, Syed Zohaib; Conci, Nicola. - (2021), pp. 8703-8710. (Intervento presentato al convegno ICPR2020 tenutosi a Milano nel 10-15 January 2021) [10.1109/ICPR48806.2021.9413077].

AG-GAN: An Attentive Group-Aware GAN for pedestrian trajectory prediction

Yue Song;Niccolò Bisagno;Syed Zohaib Hassan;Nicola Conci
2021-01-01

Abstract

Understanding human behaviors in crowded scenarios requires analyzing not only the position of the subjects in space, but also the scene context. Existing approaches mostly rely on the motion history of each pedestrian and model the interactions among people by considering the entire surrounding neighborhood. In our approach, we address the problem of motion prediction by applying coherent group clustering and a global attention mechanism on the LSTM-based Generative Adversarial Networks (GANs). The proposed model consists of an attentive group-aware GAN that observes the agents' past motion and predicts future paths, using (i) a group pooling module to model neighborhood interaction, and (ii) an attention module to specifically focus on hidden states. The experimental results demonstrate that our proposal outperforms state-of-the-art models on common benchmark datasets, and is able to generate socially-acceptable trajectories.
2021
2020 25th International Conference on Pattern Recognition (ICPR)
USA
IEEE
978-1-7281-8808-9
978-1-7281-8809-6
Song, Yue; Bisagno, Niccolò; Hassan, Syed Zohaib; Conci, Nicola
AG-GAN: An Attentive Group-Aware GAN for pedestrian trajectory prediction / Song, Yue; Bisagno, Niccolò; Hassan, Syed Zohaib; Conci, Nicola. - (2021), pp. 8703-8710. (Intervento presentato al convegno ICPR2020 tenutosi a Milano nel 10-15 January 2021) [10.1109/ICPR48806.2021.9413077].
File in questo prodotto:
File Dimensione Formato  
ICPR_2020_AGGAN.pdf

Solo gestori archivio

Tipologia: Post-print referato (Refereed author’s manuscript)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 2.38 MB
Formato Adobe PDF
2.38 MB Adobe PDF   Visualizza/Apri
AG-GAN_An_Attentive_Group-Aware_GAN_for_pedestrian_trajectory_prediction.pdf

Solo gestori archivio

Tipologia: Versione editoriale (Publisher’s layout)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 2.47 MB
Formato Adobe PDF
2.47 MB Adobe PDF   Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11572/330729
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 6
social impact