ICE: Inter-instance Contrastive Encoding for Unsupervised Person Re-identification

ICCV 2021  ·  Hao Chen, Benoit Lagadec, Francois Bremond ·

Unsupervised person re-identification (ReID) aims at learning discriminative identity features without annotations. Recently, self-supervised contrastive learning has gained increasing attention for its effectiveness in unsupervised representation learning. The main idea of instance contrastive learning is to match a same instance in different augmented views. However, the relationship between different instances has not been fully explored in previous contrastive methods, especially for instance-level contrastive loss. To address this issue, we propose Inter-instance Contrastive Encoding (ICE) that leverages inter-instance pairwise similarity scores to boost previous class-level contrastive ReID methods. We first use pairwise similarity ranking as one-hot hard pseudo labels for hard instance contrast, which aims at reducing intra-class variance. Then, we use similarity scores as soft pseudo labels to enhance the consistency between augmented and original views, which makes our model more robust to augmentation perturbations. Experiments on several large-scale person ReID datasets validate the effectiveness of our proposed unsupervised method ICE, which is competitive with even supervised methods. Code is made available at https://github.com/chenhao2345/ICE.

PDF Abstract ICCV 2021 PDF ICCV 2021 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Unsupervised Person Re-Identification DukeMTMC-reID ICE Rank-1 83.3 # 4
Rank-10 94.1 # 2
Rank-5 91.5 # 3
MAP 69.9 # 4
Unsupervised Person Re-Identification Market-1501 ICE Rank-1 93.8 # 9
MAP 82.3 # 10
Rank-10 98.4 # 8
Rank-5 97.6 # 7

Methods