1
Return

Gradually target activating and self-supervised enhancing for visual tracking

delete2026-08-03
delete0
PRE
AI
Y
Yanfang Deng
J
Jie Chen
C
Canlong Zhang *
Z
Zhixin Li
DOI：10.1016/j.knosys.2026.116779delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Designed asymmetric mixed attention to activate features and prevent template contamination. • Proposed max-suppression spatio-temporal fusion to enhance temporal cue learning capability. • Developed self-supervised decoder to guide encoder focus on target feature extraction. • Presented GFATrack, integrating these modules organically for robust visual tracking.
Keywords:
Target tracking
Feature activation
Spatio-temporal enhancement

Journal

K
Knowledge-Based Systems
IF:
7.6
Papers:
1.2W
Citations:
4.5W

Organization

G
guangxi normal university
Scholars:
1.4K
Papers: 485
Citations: 0
Cited Papers

Cited Papers

Citing Papers

Citing Papers