arrow
返回

TICNet: A Target-Insight Correlation Network for Object Tracking

delete2022-11-01
delete11
PRE
AI
W
Weijian Ruan
Mang Ye 封面图
Mang Ye (Mang Ye) *
Y
Yi Wu
刘
刘武 (Wu Liu)
J
Jun Chen
梁
梁超 (Chao Liang)
G
Ge Li
C
Chia‐Wen Lin
DOI:10.1109/TCYB.2021.3070677delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Recently, the correlation filter (CF) and Siamese network have become the two most popular frameworks in object tracking. Existing CF trackers, however, are limited by feature learning and context usage, making them sensitive to boundary effects. In contrast, Siamese trackers can easily suffer from the interference of semantic distractors. To address the above problems, we propose an end-to-end target-insight correlation network (TICNet) for object tracking, which aims at breaking the above limitations on top of a unified network. TICNet is an asymmetric dual-branch network involving a target-background awareness model (TBAM), a spatial-channel attention network (SCAN), and a distractor-aware filter (DAF) for end-to-end learning. Specifically, TBAM aims to distinguish a target from the background in the pixel level, yielding a target likelihood map based on color statistics to mine distractors for DAF learning. SCAN consists of a basic convolutional network, a channel-attention network, and a spatial-attention network, aiming to generate attentive weights to enhance the representation learning of the tracker. Especially, we formulate a differentiable DAF and employ it as a learnable layer in the network, thus helping suppress distracting regions in the background. During testing, DAF, together with TBAM, yields a response map for the final target estimation. Extensive experiments on seven benchmarks demonstrate that TICNet outperforms the state-of-the-art methods while running at real-time speed.
Keyword:
Target tracking
Correlation
Training
Task analysis
Standards
Object tracking
Visualization
Distractor-aware filter (DAF)
dual-branch network
spatial-channel attention
target-background awareness
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

IEEE Transactions on Cybernetics 封面图
IEEE Transactions on Cybernetics
IF:
10.5
论文数:
1.1W
被引数:
5.0W

机构

N
National Tsing Hua University
学者数:
1.6W
论文数: 1.4W
被引数: 1.7W
S
shenzhen institute of advanced technology, cas
学者数:
5.6K
论文数: 4.5K
被引数: 7
C
china electronics technology group
学者数:
1.8K
论文数: 1.4K
被引数: 0
P
peking university
学者数:
11.9W
论文数: 8.7W
被引数: 146
W
wuhan university
学者数:
8.1W
论文数: 5.8W
被引数: 70
学者 查看更多机构
引用论文

引用论文

Trust in Virtual Teams: A Multidisciplinary Review and Integration
err2019-01-21
err0
errOAAI
errJanine Viol Hacker; Michael Johnson; Carol Saunders; Amanda L. Thayer
err分享
err收藏
Binding of the Fluorescein Derivative Eosin Y to the Mitochondrial ADP/ATP Carrier:  Characterization of the Adenine Nucleotide Binding Site
err1998-01-06
err0
PREAI
errEiji Majima; Nana Yamaguchi; Hiroshi Chuman; Yasuo Shinohara; Mayumi Ishida; Satoru Goto; Hiroshi Terada
err分享
err收藏
err分享
err收藏
Definitions of primary-progressive multiple sclerosis trajectories by rate of clinical disability progression
err2021-05-01
err0
PREAI
errAnat Achiron; Sapir Dreyer-Alster; Michael Gurevich; Shay Menascu; David Magalashvili; Mark Dolev; Yael Stern; Tomer Ziv-Baran
err分享
err收藏
学者 查看更多内容