Exploiting Robust Unsupervised Video Person Re-identification

9 Nov 2021  ยท  Xianghao Zang, Ge Li, Wei Gao, Xiujun Shu ยท

Unsupervised video person re-identification (reID) methods usually depend on global-level features. And many supervised reID methods employed local-level features and achieved significant performance improvements. However, applying local-level features to unsupervised methods may introduce an unstable performance. To improve the performance stability for unsupervised video reID, this paper introduces a general scheme fusing part models and unsupervised learning. In this scheme, the global-level feature is divided into equal local-level feature. A local-aware module is employed to explore the poentials of local-level feature for unsupervised learning. A global-aware module is proposed to overcome the disadvantages of local-level features. Features from these two modules are fused to form a robust feature representation for each input image. This feature representation has the advantages of local-level feature without suffering from its disadvantages. Comprehensive experiments are conducted on three benchmarks, including PRID2011, iLIDS-VID, and DukeMTMC-VideoReID, and the results demonstrate that the proposed approach achieves state-of-the-art performance. Extensive ablation studies demonstrate the effectiveness and robustness of proposed scheme, local-aware module and global-aware module. The code and generated features are available at https://github.com/deropty/uPMnet.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Unsupervised Person Re-Identification DukeMTMC-VideoReID uPMnet mAP 76.9 # 2
Rank-1 83.6 # 2
Rank-5 93.1 # 1
Rank-20 97.2 # 1
Unsupervised Person Re-Identification iLIDS-VID uPMnet Rank-1 63.1 # 1
Rank-5 81.9 # 1
Rank-20 92.5 # 1
Person Re-Identification iLIDS-VID uPMnet Rank-1 63.1 # 8
Rank-20 92.5 # 6
Rank-5 81.9 # 5
Unsupervised Person Re-Identification PRID2011 uPMnet Rank-1 92.00 # 1
Rank-5 97.7 # 1
Rank-20 100.0 # 1
Person Re-Identification PRID2011 uPMnet Rank-1 92.0 # 5
Rank-20 100.0 # 1
Rank-5 97.7 # 3

Methods


No methods listed for this paper. Add relevant methods here