Journal
NEUROCOMPUTING
Volume 386, Issue -, Pages 97-109Publisher
ELSEVIER
DOI: 10.1016/j.neucom.2019.12.100
Keywords
Cross-modality person re-identification; Hetero-Center loss; Local feature
Categories
Funding
- National Natural Science Foundation of China [61501177, 61772455, 61572486, U1713213, 61902084]
- Guangdong Natural Science Foundation [2017A030310639]
- Featured Innovation Project of Guangdong Education Department [2018KTSCX174]
- Yunnan Natural Science Funds [2018FY001(-013), 2019FA-045]
- Program for Excellent Young Talents of National Natural Science Foundation of Yunnan University [2018YDJQ004]
- Program for Excellent Young Talents of Yunnan University [WX069051]
- Project of Innovative Research Team of Yunnan Province [2018HC019]
- Guangzhou University's training program for excellent new-recruited doctors [YB201712]
Ask authors/readers for more resources
Cross-modality person re-identification is a challenging problem which retrieves a given pedestrian image in RGB modality among all the gallery images in infrared modality. The task can address the limitation of RGB-based person Re-ID in dark environments. Existing researches mainly focus on enlarging inter-class differences of feature to solve the problem. However, few studies investigate improving intraclass cross-modality similarity, which is important for this issue. In this paper, we propose a novel loss function, called Hetero-Center loss (HC loss) to reduce the intra-class cross-modality variations. Specifically, HC loss can supervise the network learning the cross-modality invariant information by constraining the intra-class center distance between two heterogenous modalities. With the joint supervision of Cross-Entropy (CE) loss and HC loss, the network is trained to achieve two vital objectives, inter-class discrepancy and intra-class cross-modality similarity as much as possible. Besides, we propose a simple and high- performance network architecture to learn local feature representations for cross-modality person re-identification, which can be a baseline for future research. Extensive experiments indicate the effectiveness of the proposed methods, which outperform state-of-the-art methods by a wide margin. (c) 2019 Elsevier B.V. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available