Visible-Infrared Person Re-Identification With Real-World Label Noise
Beijing Institute of Technology · Guizhou Electromechanical Research and Design Institute · +6 more institutions
Abstract
In recent years, growing needs for advanced security and traffic management have significantly heightened the prominence of the visible-infrared person re-identification community (VI-ReID), garnering considerable attention. A critical challenge in VI-ReID is the performance degradation attributable to label noise, an issue that becomes even more pronounced in cross-modal scenarios due to an increased likelihood of data confusion. While previous methods have achieved notable successes, they often overlook the complexities of instance-dependent and real-world noise, creating a disconnect from the practical applications of person re-identification. To bridge this gap, our research analyzes the primary sources of…
Citation impact
- FWCI
- 41.75
- Percentile
- 100%
- References
- 88
Authors
6- RZRuiheng ZhangCorresponding
Beijing Institute of Technology, Guizhou Electromechanical Research and Design Institute, Beijing Jingshida Electromechanical Equipment Research Institute
- ZCZhe Cao
Beijing Institute of Technology, Guizhou Electromechanical Research and Design Institute, Beijing Jingshida Electromechanical Equipment Research Institute
- YHYan Huang
Chinese Academy of Sciences, Institute of Automation
- SYShuo Yang
Shenzhen Institute of Information Technology, Harbin Institute of Technology
- LXLixin Xu
Beijing Institute of Technology, Guizhou Electromechanical Research and Design Institute, Beijing Jingshida Electromechanical Equipment Research Institute
Topics & keywords
- Computer science
- Identification (biology)
- Noise (video)
- Infrared
- Artificial intelligence
- Computer vision
- Pattern recognition (psychology)
- Optics