Adversarial Deep Tracking

doi:10.1109/TCSVT.2018.2856540

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 图像与视频分析团队

	Adversarial Deep Tracking
	Zhao, Fei1 ; Wang, Jinqiao1 ; Wu, Yi 2,3; Tang, Ming1
刊名	IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
	2019-07-01
卷号	29 期号:7 页码:1998-2011
关键词	Visual tracking deep learning adversarial training attention
ISSN号	1051-8215
DOI	10.1109/TCSVT.2018.2856540
通讯作者	Wang, Jinqiao(jqwang@nlpr.ia.ac.cn) ; Wu, Yi(ywu.china@gmail.com)
英文摘要	A number of visual tracking methods achieve the state-of-the-art performance based on deep learning recently. However, most of these trackers utilize the deep neural network in regression task or classification task separately. In this paper, we propose an adversarial deep tracking framework. The framework is composed of a fully convolutional Siamese neural network (regression network) and a discriminative classification network. Then, we jointly optimize the regression network and the classification network by adversarial learning. In the uniform framework, the regression network and classification network can be trained end-to-end as a whole using large amounts of video training data sets. During the testing phase, the regression network generates a response map which reflects the location and the size of the target within each candidate search patch, and the classification network discriminates which response map is the best in terms of the corresponding template patch and candidate search patch. In addition, we propose an attention visualization algorithm for our tracker, and it reflects the area that attracts the attention of our tracker during tracking. The experimental results on three large-scale visual tracking benchmarks (OTB-100, TC-128, and VOT2016) demonstrate the effectiveness of the proposed tracking algorithm and show that our tracker performs comparably against the state-of-the-art trackers.
资助项目	Natural Science Foundation of China[61772527]
WOS关键词	VISUAL TRACKING ; OBJECT TRACKING ; NETWORKS
WOS研究方向	Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000473623800009
资助机构	Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/23578]
专题	自动化研究所_模式识别国家重点实验室_图像与视频分析团队
通讯作者	Wang, Jinqiao; Wu, Yi
作者单位	1.Univ Chinese Acad Sci, Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Indiana Univ Sch Med, Dept Med, Indianapolis, IN 46202 USA 3.Nanjing Audit Univ, Sch Informat Engn, Nanjing 211815, Jiangsu, Peoples R China
推荐引用方式 GB/T 7714	Zhao, Fei,Wang, Jinqiao,Wu, Yi,et al. Adversarial Deep Tracking[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,2019,29(7):1998-2011.
APA	Zhao, Fei,Wang, Jinqiao,Wu, Yi,&Tang, Ming.(2019).Adversarial Deep Tracking.IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,29(7),1998-2011.
MLA	Zhao, Fei,et al."Adversarial Deep Tracking".IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 29.7(2019):1998-2011.