RGBT Tracking by Trident Fusion Network

doi:10.1109/TCSVT.2021.3067997

CASIA OpenIR > 智能感知与计算研究中心

	RGBT Tracking by Trident Fusion Network
	Zhu, Yabin 1,2; Li, Chenglong 1,2; Tang, Jin 1,2; Luo, Bin 1,2; Wang, Liang3
发表期刊	IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
ISSN	1051-8215
	2022-02-01
卷号	32 期号:2 页码:579-592
通讯作者	Li, Chenglong(lcl1314@foxmail.com)
摘要	In recent years, RGBT tracking has become a hot topic in the field of visual tracking, and made great progress. In this paper, we propose a novel Trident Fusion Network (TFNet) to achieve effective fusion of different modalities for robust RGBT tracking. In specific, to deploy the complementarity of features of all convolutional layers, we propose a recursive strategy to densely aggregate these features that yield robust representations of target objects in two modalities. Moreover, we design a trident architecture to integrate the fused features and both modality-specific features for robust target representations. There are three main advantages. First, retaining the classification layer of each modality is beneficial to enhance feature learning of single modality, and compared with aggregate branches, single-modality branches pay more attention to the mining of modal specific information. Second, when some modality is noisy or invalid, the modality-specific branches would capture more discriminative features for RGBT tracking. Finally, the integration of aggregation branches and single-modality branches is beneficial to the complementary learning of different modalities. In addition, we also introduce a feature pruning module in each branch to prune the redundant features and avoid network overfitting. Experimental results on four RGBT tracking benchmark datasets suggest that our tracker achieves superior performance against the state-of-the-art RGBT tracking methods.
关键词	Feature extraction Convolution Target tracking Training Aggregates Visualization Benchmark testing RGBT tracking feature aggregation feature pruning trident architecture
DOI	10.1109/TCSVT.2021.3067997
关键词[WOS]	VISUAL TRACKING ; ROBUST TRACKING ; FILTER
收录类别	SCI
语种	英语
资助项目	Major Project for New Generation of AI[2018AAA0100400] ; National Natural Science Foundation of China[61976003] ; National Natural Science Foundation of China[62076003] ; National Natural Science Foundation of China[61860206004] ; Natural Science Foundation for the Higher Education Institutions of Anhui Province[KJ2020A0061] ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR)
项目资助者	Major Project for New Generation of AI ; National Natural Science Foundation of China ; Natural Science Foundation for the Higher Education Institutions of Anhui Province ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR)
WOS研究方向	Engineering
WOS类目	Engineering, Electrical & Electronic
WOS记录号	WOS:000752017700013
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：33[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/47603
专题	智能感知与计算研究中心
通讯作者	Li, Chenglong
作者单位	1.Anhui Univ, Key Lab Intelligent Comp & Signal Proc, Minist Educ, Hefei 230601, Peoples R China 2.Anhui Univ, Sch Comp Sci & Technol, Anhui Prov Key Lab Multimodal Cognit Computat, Hefei 230601, Peoples R China 3.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Zhu, Yabin,Li, Chenglong,Tang, Jin,et al. RGBT Tracking by Trident Fusion Network[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,2022,32(2):579-592.
APA	Zhu, Yabin,Li, Chenglong,Tang, Jin,Luo, Bin,&Wang, Liang.(2022).RGBT Tracking by Trident Fusion Network.IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,32(2),579-592.
MLA	Zhu, Yabin,et al."RGBT Tracking by Trident Fusion Network".IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 32.2(2022):579-592.