Home »
Academic Articles »
Art »
Part‐level attention networks for cross‐domain person re‐identification
Part‐level attention networks for cross‐domain person re‐identification
Qun Zhao
(College of Big Data and Information Engineering Guizhou University Guiyang Guizhou China)
Nisuo Du
(Guizhou Provincial Key Laboratory of Public Big Data Guiyang Guizhou China)
Zhi Ouyang
(Guizhou Provincial Key Laboratory of Public Big Data Guiyang Guizhou China)
Ning Kang
(University of Electronic Science and Technology of China Chengdu China)
Ziyan Liu
(College of Big Data and Information Engineering Guizhou University Guiyang Guizhou China)
Xu Wang
(College of Big Data and Information Engineering Guizhou University Guiyang Guizhou China)
Qing He
(College of Big Data and Information Engineering Guizhou University Guiyang Guizhou China)
Yiling Xu
(Guizhou Institute of Quantum Information and Big Data Applied Technology Guiyang Guizhou China)
Shichun Ge
(Guizhou Institute of Quantum Information and Big Data Applied Technology Guiyang Guizhou China)
Jingkuan Song
(University of Electronic Science and Technology of China Chengdu China)
Journal: IET Image Processing
•
• Wiley
• ISSN: 1751-9667
Abstract
Abstract Person re‐identification (Re‐ID) is in significant demand for intelligent security and single or multiple‐target tracking. However, there are issues in the person Re‐ID tasks, such as sharp decline in cross‐data sets detection accuracy, poor generalization and cross‐domain ability of the model. This work mainly studies the generalization and adaptation of cross‐domain person Re‐ID models. Different from most existing methods for cross‐domain Re‐ID tasks, the authors use diversified spatial semantic feature in pixel‐level learning in the target domain to improve the generality and adaptability of the model. In the case that no information of the target domain is used during the model training, the trained model is directly tested on the data set of the target domain. It has proven effective to add the attention cascade module into the backbone network combining with the part‐level branch. The authors conducted extensive experiments based on the three data sets of Market‐1501, DukeMTMC‐ReID and MSMT17, resulting in both single‐domain and cross‐domain tests with an average improvement of Rank1 and mAP values of about 10% compared with Baseline through the authors' proposed method named Part‐Level Attention Network.
Qun Zhao, Nisuo Du, Zhi Ouyang, et al. (2021). "Part‐level attention networks for cross‐domain person re‐identification". IET Image Processing. https://doi.org/10.1049/ipr2.12292