Weakly Supervised Person Search with Region Siamese Networks

Han, C; Su, K; Yu, D; Yuan, Z; Gao, C; Sang, N; Yang, Y; Wang, C

Weakly Supervised Person Search with Region Siamese Networks

Han, C Su, K Yu, D Yuan, Z Gao, C Sang, N Yang, Y

Wang, C

Permalink

Publisher:: IEEE
Publication Type:: Conference Proceeding
Citation:: 2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2022, 00, pp. 11986-11995
Issue Date:: 2022-02-28

Closed Access

	Filename	Description	Size
	Weakly Supervised Person Search with Region Siamese Networks.pdf	Published version	5.2 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	Han, C
dc.contributor.author	Su, K
dc.contributor.author	Yu, D
dc.contributor.author	Yuan, Z
dc.contributor.author	Gao, C
dc.contributor.author	Sang, N
dc.contributor.author	Yang, Y https://orcid.org/0000-0002-0512-880X
dc.contributor.author	Wang, C
dc.date	2021-10-10
dc.date.accessioned	2023-04-06T05:05:51Z
dc.date.available	2023-04-06T05:05:51Z
dc.date.issued	2022-02-28
dc.identifier.citation	2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2022, 00, pp. 11986-11995
dc.identifier.isbn	9781665428125
dc.identifier.issn	1550-5499
dc.identifier.uri	http://hdl.handle.net/10453/169336
dc.description.abstract	Supervised learning is dominant in person search, but it requires elaborate labeling of bounding boxes and identities. Large-scale labeled training data is often difficult to collect, especially for person identities. A natural question is whether a good person search model can be trained without the need of identity supervision. In this paper, we present a weakly supervised setting where only bounding box annotations are available. Based on this new setting, we provide an effective baseline model termed Region Siamese Networks (R-SiamNets). Towards learning useful representations for recognition in the absence of identity labels, we supervise the R-SiamNet with instance-level consistency loss and cluster-level contrastive loss. For instance-level consistency learning, the R-SiamNet is constrained to extract consistent features from each person region with or without out-of-region context. For cluster-level contrastive learning, we enforce the aggregation of closest instances and the separation of dissimilar ones in feature space. Extensive experiments validate the utility of our weakly supervised method. Our model achieves the rank-1 of 87.1% and mAP of 86.0% on CUHK-SYSU benchmark, which surpasses several fully supervised methods, such as OIM [36] and MGTS [4], by a clear margin. More promising performance can be reached by incorporating extra training data. We hope this work could encourage the future research in this field.
dc.language	en
dc.publisher	IEEE
dc.relation.ispartof	2021 IEEE/CVF International Conference on Computer Vision (ICCV)
dc.relation.ispartof	2021 IEEE/CVF International Conference on Computer Vision
dc.relation.isbasedon	10.1109/iccv48922.2021.01179
dc.rights	info:eu-repo/semantics/closedAccess
dc.title	Weakly Supervised Person Search with Region Siamese Networks
dc.type	Conference Proceeding
utslib.citation.volume	00
utslib.location.activity	Montreal, QC, Canada
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
utslib.copyright.status	closed_access	*
dc.date.updated	2023-04-06T05:05:49Z
pubs.finish-date	2021-10-17
pubs.publication-status	Published
pubs.start-date	2021-10-10
pubs.volume	00

Abstract:

Supervised learning is dominant in person search, but it requires elaborate labeling of bounding boxes and identities. Large-scale labeled training data is often difficult to collect, especially for person identities. A natural question is whether a good person search model can be trained without the need of identity supervision. In this paper, we present a weakly supervised setting where only bounding box annotations are available. Based on this new setting, we provide an effective baseline model termed Region Siamese Networks (R-SiamNets). Towards learning useful representations for recognition in the absence of identity labels, we supervise the R-SiamNet with instance-level consistency loss and cluster-level contrastive loss. For instance-level consistency learning, the R-SiamNet is constrained to extract consistent features from each person region with or without out-of-region context. For cluster-level contrastive learning, we enforce the aggregation of closest instances and the separation of dissimilar ones in feature space. Extensive experiments validate the utility of our weakly supervised method. Our model achieves the rank-1 of 87.1% and mAP of 86.0% on CUHK-SYSU benchmark, which surpasses several fully supervised methods, such as OIM [36] and MGTS [4], by a clear margin. More promising performance can be reached by incorporating extra training data. We hope this work could encourage the future research in this field.

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

http://hdl.handle.net/10453/169336