Research Article
Research on Multi-scale Pedestrian Attribute Recognition Based on Dual Self-attention Mechanism
@INPROCEEDINGS{10.1007/978-3-031-55471-1_16, author={He Xiao and Wenbiao Xie and Yang Zhou and Yong Luo and Ruoni Zhang and Xiao Xu}, title={Research on Multi-scale Pedestrian Attribute Recognition Based on Dual Self-attention Mechanism}, proceedings={Mobile Networks and Management. 13th EAI International Conference, MONAMI 2023, Yingtan, China, October 27-29, 2023, Proceedings}, proceedings_a={MONAMI}, year={2024}, month={3}, keywords={incomplete feature dual self-attention multi-scale fusion pedestrian attribute recognition}, doi={10.1007/978-3-031-55471-1_16} }- He Xiao
Wenbiao Xie
Yang Zhou
Yong Luo
Ruoni Zhang
Xiao Xu
Year: 2024
Research on Multi-scale Pedestrian Attribute Recognition Based on Dual Self-attention Mechanism
MONAMI
Springer
DOI: 10.1007/978-3-031-55471-1_16
Abstract
As one of the important fields of computer vision research, pedestrian attribute recognition has gained increasing attention from domestic and foreign researchers due to its huge potential applications. However, obtaining long-distance pedestrian information in actual scenes poses challenges such as lack of information, incomplete feature extraction, and low attribute recognition accuracy. To address these issues, we propose a multi-scale feature fusion network based on a dual self-attention mechanism. The fusion module merges multi-scale features to enable more complete attribute extraction, while the dual self-attention module focuses the network on important regions. Experimental results on PA-100K, RAP, and PETA datasets achieved mean accuracies of 81.97%, 81.53%, and 86.37%, respectively. Extensive experiments demonstrate that the proposed method is highly competitive in pedestrian attribute recognition.
