In this paper, we address the task of semantic-guided scene generation. One open challenge widely observed in global image-level generation methods is the difficulty of generating small objects and detailed local texture. To tackle this issue, in this work we consider learning the scene generation in a local context, and correspondingly design a local class-specific generative network with semantic maps as a guidance, which separately constructs and learns sub-generators concentrating on the generation of different classes, and is able to provide more scene details. To learn more discriminative class-specific feature representations for the local generation, a novel classification module is also proposed. To combine the advantage of both global image-level and the local class-specific generation, a joint generation network is designed with an attention fusion module and a dual-discriminator structure embedded. Extensive experiments on two scene image generation tasks show superior generation performance of the proposed model. State-of-the-art results are established by large margins on both tasks and on challenging public benchmarks. The source code and trained models are available at https://github.com/Ha0Tang/LGGAN.

Local Class-Specific and Global Image-Level Generative Adversarial Networks for Semantic-Guided Scene Generation / Tang, Hao; Xu, Dan; Yan, Yan; Torr, Philip H. S.; Sebe, Nicu. - (2020), pp. 7867-7876. ((Intervento presentato al convegno IEEE CVPR 2020 tenutosi a Virtual nel 14th-19th June 2020 [10.1109/CVPR42600.2020.00789].

Local Class-Specific and Global Image-Level Generative Adversarial Networks for Semantic-Guided Scene Generation

Tang, Hao;Xu, Dan;Yan, Yan;Sebe, Nicu
2020

Abstract

In this paper, we address the task of semantic-guided scene generation. One open challenge widely observed in global image-level generation methods is the difficulty of generating small objects and detailed local texture. To tackle this issue, in this work we consider learning the scene generation in a local context, and correspondingly design a local class-specific generative network with semantic maps as a guidance, which separately constructs and learns sub-generators concentrating on the generation of different classes, and is able to provide more scene details. To learn more discriminative class-specific feature representations for the local generation, a novel classification module is also proposed. To combine the advantage of both global image-level and the local class-specific generation, a joint generation network is designed with an attention fusion module and a dual-discriminator structure embedded. Extensive experiments on two scene image generation tasks show superior generation performance of the proposed model. State-of-the-art results are established by large margins on both tasks and on challenging public benchmarks. The source code and trained models are available at https://github.com/Ha0Tang/LGGAN.
Proceedings: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition
Piscataway, NJ
IEEE Computer Society
978-1-7281-7168-5
Tang, Hao; Xu, Dan; Yan, Yan; Torr, Philip H. S.; Sebe, Nicu
Local Class-Specific and Global Image-Level Generative Adversarial Networks for Semantic-Guided Scene Generation / Tang, Hao; Xu, Dan; Yan, Yan; Torr, Philip H. S.; Sebe, Nicu. - (2020), pp. 7867-7876. ((Intervento presentato al convegno IEEE CVPR 2020 tenutosi a Virtual nel 14th-19th June 2020 [10.1109/CVPR42600.2020.00789].
File in questo prodotto:
File Dimensione Formato  
Tang_Local_Class-Specific_and_Global_Image-Level_Generative_Adversarial_Networks_for_Semantic-Guided_CVPR_2020_paper.pdf

accesso aperto

Tipologia: Post-print referato (Refereed author’s manuscript)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 2.84 MB
Formato Adobe PDF
2.84 MB Adobe PDF Visualizza/Apri
Local_Class-Specific_and_Global_Image-Level_Generative_Adversarial_Networks_for_Semantic-Guided_Scene_Generation.pdf

Solo gestori archivio

Tipologia: Versione editoriale (Publisher’s layout)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 2.31 MB
Formato Adobe PDF
2.31 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: http://hdl.handle.net/11572/284530
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 38
  • ???jsp.display-item.citation.isi??? ND
social impact