Transformer Based Multi-Grained Features for Unsupervised Person Re-Identification

22 Nov 2022  ·  Jiachen Li, Menglin Wang, Xiaojin Gong ·

Multi-grained features extracted from convolutional neural networks (CNNs) have demonstrated their strong discrimination ability in supervised person re-identification (Re-ID) tasks. Inspired by them, this work investigates the way of extracting multi-grained features from a pure transformer network to address the unsupervised Re-ID problem that is label-free but much more challenging. To this end, we build a dual-branch network architecture based upon a modified Vision Transformer (ViT). The local tokens output in each branch are reshaped and then uniformly partitioned into multiple stripes to generate part-level features, while the global tokens of two branches are averaged to produce a global feature. Further, based upon offline-online associated camera-aware proxies (O2CAP) that is a top-performing unsupervised Re-ID method, we define offline and online contrastive learning losses with respect to both global and part-level features to conduct unsupervised learning. Extensive experiments on three person Re-ID datasets show that the proposed method outperforms state-of-the-art unsupervised methods by a considerable margin, greatly mitigating the gap to supervised counterparts. Code will be available soon at https://github.com/RikoLi/WACV23-workshop-TMGF.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Unsupervised Person Re-Identification DukeMTMC-reID TMGF Rank-1 86.7 # 1
Rank-10 94.1 # 2
Rank-5 92.9 # 1
MAP 76.8 # 1
Unsupervised Person Re-Identification Market-1501 TMGF Rank-1 95.5 # 1
MAP 89.5 # 2
Rank-10 98.7 # 1
Rank-5 98.0 # 1
Unsupervised Person Re-Identification MSMT17 TMGF mAP 58.2 # 2
Rank-1 83.3 # 2
Rank-5 90.2 # 2
Rank-10 92.1 # 2

Methods