Improving person re-identification by attribute and identity learning

Authors: Yutian Lin, Liang Zheng, Zhedong Zheng, Yu Wu, Zhilan Hu, Chenggang Yan, Yi Yang

Published in Pattern Recognition (PR), 2019

Recommended citation: Yutian Lin, Liang Zheng, Zhedong Zheng, Yu Wu, Zhilan Hu, Chenggang Yan, Yi Yang, "Improving person re-identification by attribute and identity learning." Pattern Recognition, 2019. DOI: 10.1016/j.patcog.2019.06.006
Download PDF: https://zdzheng.xyz/files/2019/PR19.pdf

Code is available at: https://github.com/vana77/Market-1501_Attribute

Abstract: Person re-identification (re-ID) and attribute recognition share a common target at learning pedestrian descriptions. Their difference consists in the granularity. Most existing re-ID methods only take identity labels of pedestrians into consideration. However, we find the attributes, containing detailed local descriptions, are beneficial in allowing the re-ID model to learn more discriminative feature representations. In this paper, based on the complementarity of attribute labels and ID labels, we propose an attribute-person recognition (APR) network, a multi-task network which learns a re-ID embedding and at the same time predicts pedestrian attributes. We manually annotate attribute labels for two large-scale re-ID datasets, and systematically investigate how person re-ID and attribute recognition benefit from each other. In addition, we re-weight the attribute predictions considering the dependencies and correlations among the attributes. The experimental results on two large-scale re-ID benchmarks demonstrate that by learning a more discriminative representation, APR achieves competitive re-ID performance compared with the state-of-the-art methods. We use APR to speed up the retrieval process by ten times with a minor accuracy drop of 2.92\% on Market-1501. Besides, we also apply APR on the attribute recognition task and demonstrate improvement over the baselines.

@article{lin2019improving,
author = "Lin, Yutian and Zheng, Liang and Zheng, Zhedong and Wu, Yu and Hu, Zhilan and Yan, Chenggang and Yang, Yi",
doi = "10.1016/j.patcog.2019.06.006",
title = "Improving person re-identification by attribute and identity learning",
abstract = "Person re-identification (re-ID) and attribute recognition share a common target at learning pedestrian descriptions. Their difference consists in the granularity. Most existing re-ID methods only take identity labels of pedestrians into consideration. However, we find the attributes, containing detailed local descriptions, are beneficial in allowing the re-ID model to learn more discriminative feature representations. In this paper, based on the complementarity of attribute labels and ID labels, we propose an attribute-person recognition (APR) network, a multi-task network which learns a re-ID embedding and at the same time predicts pedestrian attributes. We manually annotate attribute labels for two large-scale re-ID datasets, and systematically investigate how person re-ID and attribute recognition benefit from each other. In addition, we re-weight the attribute predictions considering the dependencies and correlations among the attributes. The experimental results on two large-scale re-ID benchmarks demonstrate that by learning a more discriminative representation, APR achieves competitive re-ID performance compared with the state-of-the-art methods. We use APR to speed up the retrieval process by ten times with a minor accuracy drop of 2.92\\% on Market-1501. Besides, we also apply APR on the attribute recognition task and demonstrate improvement over the baselines.",
journal = "Pattern Recognition",
volume = "95",
pages = "151--161",
year = "2019",
url = "https://zdzheng.xyz/files/2019/PR19.pdf",
code = "https://github.com/vana77/Market-1501\_Attribute",
publisher = "Elsevier" }