Rethinking Sampling Strategies for Unsupervised Person Re-identification
This work addresses a specific bottleneck in unsupervised person re-identification for computer vision applications, offering an incremental improvement in sampling strategy.
The paper tackles the problem of performance differences in unsupervised person re-identification by identifying deteriorated over-fitting as a key issue and proposes a group sampling strategy to enhance statistical stability, achieving results comparable to state-of-the-art methods on datasets like Market-1501 and DukeMTMC-reID.
Unsupervised person re-identification (re-ID) remains a challenging task. While extensive research has focused on the framework design and loss function, this paper shows that sampling strategy plays an equally important role. We analyze the reasons for the performance differences between various sampling strategies under the same framework and loss function. We suggest that deteriorated over-fitting is an important factor causing poor performance, and enhancing statistical stability can rectify this problem. Inspired by that, a simple yet effective approach is proposed, termed group sampling, which gathers samples from the same class into groups. The model is thereby trained using normalized group samples, which helps alleviate the negative impact of individual samples. Group sampling updates the pipeline of pseudo-label generation by guaranteeing that samples are more efficiently classified into the correct classes. It regulates the representation learning process, enhancing statistical stability for feature representation in a progressive fashion. Extensive experiments on Market-1501, DukeMTMC-reID and MSMT17 show that group sampling achieves performance comparable to state-of-the-art methods and outperforms the current techniques under purely camera-agnostic settings. Code has been available at https://github.com/ucas-vg/GroupSampling.