Hierarchical Clustering With Hard-Batch Triplet Loss for Person Re-Identification

By Kaiwei Zeng, Munan Ning, Yaohua Wang, Yang Guo. Computer Vision and Pattern Recognition, 2019

310 citations recorded by the source corpus.

For clustering-guided fully unsupervised person reidentification (re-ID) methods, the quality of pseudo labels generated by clustering directly decides the model performance. In order to improve the quality of pseudo labels in existing methods, we propose the HCT method which combines hierarchical clustering with hard-batch triplet loss. The key idea of HCT is to make full use of the similarity among samples in the target dataset through hierarchical clustering, reduce the influence of hard examples through hard-batch triplet loss, so as to generate high quality pseudo labels and improve model performance. Specifically, (1) we use hierarchical clustering to generate pseudo labels, (2) we use PK sampling in each iteration to generate a new dataset for training, (3) we conduct training with hard-batch triplet loss and evaluate model performance in each iteration. We evaluate our model on Market-1501 and DukeMTMC-reID. Results show that HCT achieves 56.4% mAP on Market-1501 and 50.7% mAP on DukeMTMC-reID which surpasses state-of-the-arts a lot in fully unsupervised re-ID and even better than most unsupervised domain adaptation (UDA) methods which use the labeled source dataset. Code will be released soon on https://github.com/zengkaiwei/HCT

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/5bd8c74d52230c04e83da6e828d89bdfea6dd38d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

1. Shiliang Zhang
2. Dapeng Chen
3. Rui Zhao
4. Yixiao Ge
5. Feng Zhu
6. Dongkai Wang
7. In-So Kweon
8. Hongsheng Li

In Computer Vision and Pattern Recognition

1. Hongsheng Li
2. Yi Yang
3. Xiaogang Wang
4. Shiliang Zhang
5. Liang Zheng
6. Dapeng Chen
7. Qi Tian
8. In-So Kweon

Citations in the corpus

• Unsupervised Person Re-Identification via Multi-Label Classification (2020, background): “HCT (Zeng et al., 2020) addresses the first issue by re-clustering at each epoch, thus gets better performance.”
• Dual Temperature Helps Contrastive Learning Without Many Negative Samples: Towards Understanding and Simplifying MoCo (2022, background): “A key property of CL is that it involves negative samples, and a major line works [3,10,26,28,29,36,40,46,49,53,61] have shown that mining hard negative samples can be beneficial for performance.”
• Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID (2020, background): “Our unsupervised version without labeled source-domain data on object re-ID task significantly outperforms state-of-the-arts [26, 45, 53] by 16.”
• Comprehensive survey on hierarchical clustering algorithms and the recent developments (2022, methodology): “In the paper of Zeng et al. (2020), the method HCT combines the hierarchical clustering with hard-batch triplet loss for Person re-identification, the hierarchical clustering method is used to generate the pseudo labels by iteratively agglomerating two samples according to their minimal unweighted…”