Exponential Information Bottleneck Theory Against Intra-Attribute Variations for Pedestrian Attribute Recognition

Wu, J; Huang, Y; Gao, M; Gao, Z; Zhao, J; Shi, J; Zhang, A

Exponential Information Bottleneck Theory Against Intra-Attribute Variations for Pedestrian Attribute Recognition

Wu, J Huang, Y

Gao, M Gao, Z Zhao, J Shi, J Zhang, A

Permalink

Publisher:: IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Publication Type:: Journal Article
Citation:: IEEE Transactions on Information Forensics and Security, 2023, 18, pp. 5623-5635
Issue Date:: 2023-01-01

Closed Access

	Filename	Description	Size
	Exponential_Information_Bottleneck_Theory_Against_Intra-Attribute_Variations_for_Pedestrian_Attribute_Recognition.pdf	Published version	1.4 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Wu, J
dc.contributor.author	Huang, Y https://orcid.org/0000-0002-1363-5318
dc.contributor.author	Gao, M
dc.contributor.author	Gao, Z
dc.contributor.author	Zhao, J
dc.contributor.author	Shi, J
dc.contributor.author	Zhang, A
dc.date.accessioned	2024-06-20T22:55:48Z
dc.date.available	2024-06-20T22:55:48Z
dc.date.issued	2023-01-01
dc.identifier.citation	IEEE Transactions on Information Forensics and Security, 2023, 18, pp. 5623-5635
dc.identifier.issn	1556-6013
dc.identifier.issn	1556-6021
dc.identifier.uri	http://hdl.handle.net/10453/179597
dc.description.abstract	Multi-label pedestrian attribute recognition (PAR) involves assigning multiple attributes to pedestrian images captured by video surveillance cameras. Despite its importance, learning robust attribute-related features for PAR remains a challenge due to the large intra-attribute variations in the image space. These variations, which stem from changes in pedestrian poses, illumination conditions, and background noise, make extracted attribute-related features susceptible to irrelevant information or noise interference. Existing PAR methods rely on body prior extractors or attention mechanisms to locate attribute-correlation regions for extracting robust features. However, these methods may not be robust to intra-attribute variations, which limits their effectiveness. To address this challenge, we propose a novel and flexible PAR framework that leverages the exponential information bottleneck (ExpIB) approach. Our ExpIB-Net uses mutual information compression as the main penalty during the early stage of training, thereby eliminating irrelevant information. As training progresses, the mutual information penalty weakens and the Binary Cross-Entropy Loss (BCELoss) contributes to improving the PAR recognition accuracy. Our method can also be integrated into an attention module to form the AttExpIB-Net, which better handles intra-attribute variations for better performance. Additionally, our model-agnostic ExpIB approach is plug-and-play, requiring no additional computational overhead during inference. Experiments on several challenging PAR datasets show that our method outperforms state-of-the-art approaches.
dc.language	English
dc.publisher	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
dc.relation.ispartof	IEEE Transactions on Information Forensics and Security
dc.relation.isbasedon	10.1109/TIFS.2023.3311584
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	08 Information and Computing Sciences, 09 Engineering
dc.subject.classification	Strategic, Defence & Security Studies
dc.subject.classification	40 Engineering
dc.subject.classification	46 Information and computing sciences
dc.title	Exponential Information Bottleneck Theory Against Intra-Attribute Variations for Pedestrian Attribute Recognition
dc.type	Journal Article
utslib.citation.volume	18
utslib.for	08 Information and Computing Sciences
utslib.for	09 Engineering
pubs.organisational-group	University of Technology Sydney
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
utslib.copyright.status	closed_access	*
dc.date.updated	2024-06-20T22:55:46Z
pubs.publication-status	Published
pubs.volume	18

Abstract:

Multi-label pedestrian attribute recognition (PAR) involves assigning multiple attributes to pedestrian images captured by video surveillance cameras. Despite its importance, learning robust attribute-related features for PAR remains a challenge due to the large intra-attribute variations in the image space. These variations, which stem from changes in pedestrian poses, illumination conditions, and background noise, make extracted attribute-related features susceptible to irrelevant information or noise interference. Existing PAR methods rely on body prior extractors or attention mechanisms to locate attribute-correlation regions for extracting robust features. However, these methods may not be robust to intra-attribute variations, which limits their effectiveness. To address this challenge, we propose a novel and flexible PAR framework that leverages the exponential information bottleneck (ExpIB) approach. Our ExpIB-Net uses mutual information compression as the main penalty during the early stage of training, thereby eliminating irrelevant information. As training progresses, the mutual information penalty weakens and the Binary Cross-Entropy Loss (BCELoss) contributes to improving the PAR recognition accuracy. Our method can also be integrated into an attention module to form the AttExpIB-Net, which better handles intra-attribute variations for better performance. Additionally, our model-agnostic ExpIB approach is plug-and-play, requiring no additional computational overhead during inference. Experiments on several challenging PAR datasets show that our method outperforms state-of-the-art approaches.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/179597