Unsupervised person re-identification: Clustering and fine-tuning

Fan, H; Zheng, L; Yan, C; Yang, Y

Unsupervised person re-identification: Clustering and fine-tuning

Fan, H Zheng, L

Yan, C Yang, Y

Permalink

Publication Type:: Journal Article
Citation:: ACM Transactions on Multimedia Computing, Communications and Applications, 2018, 14 (4)
Issue Date:: 2018-11-01

Closed Access

	Filename	Description	Size
	a83-fan.pdf	Published Version	6.41 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	Fan, H	en_US
dc.contributor.author	Zheng, L https://orcid.org/0000-0002-1464-9500	en_US
dc.contributor.author	Yan, C	en_US
dc.contributor.author	Yang, Y https://orcid.org/0000-0001-5528-0546	en_US
dc.date.issued	2018-11-01	en_US
dc.identifier.citation	ACM Transactions on Multimedia Computing, Communications and Applications, 2018, 14 (4)	en_US
dc.identifier.issn	1551-6857	en_US
dc.identifier.uri	http://hdl.handle.net/10453/131430
dc.description.abstract	© 2018 Association for Computing Machinery. The superiority of deeply learned pedestrian representations has been reported in very recent literature of person re-identification (re-ID). In this article, we consider the more pragmatic issue of learning a deep feature with no or only a few labels. We propose a progressive unsupervised learning (PUL) method to transfer pretrained deep representations to unseen domains. Our method is easy to implement and can be viewed as an effective baseline for unsupervised re-ID feature learning. Specifically, PUL iterates between (1) pedestrian clustering and (2) fine-tuning of the convolutional neural network (CNN) to improve the initialization model trained on the irrelevant labeled dataset. Since the clustering results can be very noisy, we add a selection operation between the clustering and fine-tuning. At the beginning, when the model is weak, CNN is fine-tuned on a small amount of reliable examples that locate near to cluster centroids in the feature space. As the model becomes stronger, in subsequent iterations, more images are being adaptively selected as CNN training samples. Progressively, pedestrian clustering and the CNN model are improved simultaneously until algorithm convergence. This process is naturally formulated as self-paced learning. We then point out promising directions that may lead to further improvement. Extensive experiments on three large-scale re-ID datasets demonstrate that PUL outputs discriminative features that improve the re-ID accuracy. Our code has been released at https://github.com/hehefan/Unsupervised-Person-Re-identification-Clustering-and-Fine-tuning.	en_US
dc.relation.ispartof	ACM Transactions on Multimedia Computing, Communications and Applications	en_US
dc.relation.isbasedon	10.1145/3243316	en_US
dc.subject.classification	Artificial Intelligence & Image Processing	en_US
dc.title	Unsupervised person re-identification: Clustering and fine-tuning	en_US
dc.type	Journal Article
utslib.citation.volume	4	en_US
utslib.citation.volume	14	en_US
utslib.for	0803 Computer Software	en_US
utslib.for	0806 Information Systems	en_US
utslib.for	0805 Distributed Computing	en_US
pubs.embargo.period	Not known	en_US
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 Software
pubs.organisational-group	/University of Technology Sydney/Strength - CAI - Centre for Artificial Intelligence
pubs.organisational-group	/University of Technology Sydney/Students
utslib.copyright.status	closed_access
pubs.issue	4	en_US
pubs.publication-status	Published	en_US
pubs.volume	14	en_US

Abstract:

© 2018 Association for Computing Machinery. The superiority of deeply learned pedestrian representations has been reported in very recent literature of person re-identification (re-ID). In this article, we consider the more pragmatic issue of learning a deep feature with no or only a few labels. We propose a progressive unsupervised learning (PUL) method to transfer pretrained deep representations to unseen domains. Our method is easy to implement and can be viewed as an effective baseline for unsupervised re-ID feature learning. Specifically, PUL iterates between (1) pedestrian clustering and (2) fine-tuning of the convolutional neural network (CNN) to improve the initialization model trained on the irrelevant labeled dataset. Since the clustering results can be very noisy, we add a selection operation between the clustering and fine-tuning. At the beginning, when the model is weak, CNN is fine-tuned on a small amount of reliable examples that locate near to cluster centroids in the feature space. As the model becomes stronger, in subsequent iterations, more images are being adaptively selected as CNN training samples. Progressively, pedestrian clustering and the CNN model are improved simultaneously until algorithm convergence. This process is naturally formulated as self-paced learning. We then point out promising directions that may lead to further improvement. Extensive experiments on three large-scale re-ID datasets demonstrate that PUL outputs discriminative features that improve the re-ID accuracy. Our code has been released at https://github.com/hehefan/Unsupervised-Person-Re-identification-Clustering-and-Fine-tuning.

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

http://hdl.handle.net/10453/131430