MeMu: Metric correlation Siamese network and multi-class negative sampling for visual tracking

Xiao, Y; Li, J; Du, B; Wu, J; Chang, J; Zhang, W

MeMu: Metric correlation Siamese network and multi-class negative sampling for visual tracking

Xiao, Y Li, J Du, B Wu, J Chang, J Zhang, W

Permalink

Publisher:: Elsevier BV
Publication Type:: Journal Article
Citation:: Pattern Recognition, 2020, 100, pp. 107170-107170
Issue Date:: 2020-04-01

Closed Access

	Filename	Description	Size
	1-s2.0-S0031320319304704-main.pdf	Published version	5.72 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	Xiao, Y
dc.contributor.author	Li, J
dc.contributor.author	Du, B
dc.contributor.author	Wu, J
dc.contributor.author	Chang, J
dc.contributor.author	Zhang, W
dc.date.accessioned	2021-03-01T20:55:06Z
dc.date.available	2021-03-01T20:55:06Z
dc.date.issued	2020-04-01
dc.identifier.citation	Pattern Recognition, 2020, 100, pp. 107170-107170
dc.identifier.issn	0031-3203
dc.identifier.uri	http://hdl.handle.net/10453/146586
dc.description.abstract	© 2019 Elsevier Ltd Despite the great success in the computer vision field, visual tracking is still a challenging task. The main obstacle is that the target object often suffers from interference, such as occlusion. As most Siamese network-based trackers mainly sample image patches of target objects for training, the tracking algorithm lacks sufficient information about the surrounding environment. Besides, many Siamese network-based tracking algorithms build a regression only with the target object samples without considering the relationship between target and background, which may deteriorate the performance of trackers. In this paper, we propose a metric correlation Siamese network and multi-class negative sampling tracking method. For the first time, we explore a sampling approach that includes three different kinds of negative samples: virtual negative samples for pre-learning the potential occlusion situation, boundary negative samples to cope with potential tracking drift, and context negative samples to cope with potential incorrect positioning. With the three kinds of negative samples, we also propose a metric correlation method to train a correlation filter that contains metric information for better discrimination. Furthermore, we design a Siamese network-based architecture to embed the metric correlation filter module mentioned above in order to benefit from the powerful representation ability of deep learning. Extensive experiments on challenging OTB100 and VOT2017 datasets demonstrate the competitive performance of the proposed algorithm performs favorably compared with state-of-the-art approaches.
dc.language	en
dc.publisher	Elsevier BV
dc.relation.ispartof	Pattern Recognition
dc.relation.isbasedon	10.1016/j.patcog.2019.107170
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0801 Artificial Intelligence and Image Processing, 0806 Information Systems, 0906 Electrical and Electronic Engineering
dc.subject.classification	Artificial Intelligence & Image Processing
dc.title	MeMu: Metric correlation Siamese network and multi-class negative sampling for visual tracking
dc.type	Journal Article
utslib.citation.volume	100
utslib.for	0801 Artificial Intelligence and Image Processing
utslib.for	0806 Information Systems
utslib.for	0906 Electrical and Electronic Engineering
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney
utslib.copyright.status	closed_access	*
dc.date.updated	2021-03-01T20:55:02Z
pubs.publication-status	Published
pubs.volume	100

Abstract:

© 2019 Elsevier Ltd Despite the great success in the computer vision field, visual tracking is still a challenging task. The main obstacle is that the target object often suffers from interference, such as occlusion. As most Siamese network-based trackers mainly sample image patches of target objects for training, the tracking algorithm lacks sufficient information about the surrounding environment. Besides, many Siamese network-based tracking algorithms build a regression only with the target object samples without considering the relationship between target and background, which may deteriorate the performance of trackers. In this paper, we propose a metric correlation Siamese network and multi-class negative sampling tracking method. For the first time, we explore a sampling approach that includes three different kinds of negative samples: virtual negative samples for pre-learning the potential occlusion situation, boundary negative samples to cope with potential tracking drift, and context negative samples to cope with potential incorrect positioning. With the three kinds of negative samples, we also propose a metric correlation method to train a correlation filter that contains metric information for better discrimination. Furthermore, we design a Siamese network-based architecture to embed the metric correlation filter module mentioned above in order to benefit from the powerful representation ability of deep learning. Extensive experiments on challenging OTB100 and VOT2017 datasets demonstrate the competitive performance of the proposed algorithm performs favorably compared with state-of-the-art approaches.

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

http://hdl.handle.net/10453/146586