Camera Bias Regularization for Person Re-identification

Tao He, Tongkun Xu, Weihua Chen, Yuchen Guo, Guiguang Ding, Zhenhua Guo

Keywords: Person Re-identification, Image Retrieval

Abstract: Person re-identification (Re-ID) is to match persons captured by non-overlapping cameras. Due to the discrepancies between cameras caused by illumination, background, or viewpoint, the underlying difficulty for Re-ID is the camera bias problem, which leads to the large gap of within-identity features from different cameras. With limited cross-camera annotated data, Re-ID models tend to learn camera-related features, instead of identity-related features. Consequently, Re-ID models suffer from poor transfer ability from seen domains to unseen domains. In this paper, we investigate the camera bias problem in supervised learning, unsupervised learning, and their variants. In particular, we propose a novel Camera Bias Regularization (CBR) term to reduce the feature distribution gap between cameras by enlarging the intra-camera distance and reducing the inter-camera distance simultaneously. Extensive experiments on person Re-ID tasks validate the effectiveness and universality of the proposed CBR.

One-sentence Summary: We propose a Camera Bias Regularization (CBR) term to improve the generalization and transfer ability of Re-ID models by reducing the feature distribution gap between different cameras.