This paper addresses the problem of handling spatial misalignments due to camera-view changes or human-pose variations in person re-identification. We first introduce a boosting-based approach to learn a correspondence structure which indicates the patch-wise matching probabilities between images from a target camera pair. The learned correspondence structure can not only capture the spatial correspondence pattern between cameras but also handle the viewpoint or human-pose variation in individual images. We further introduce a global-based matching process. It integrates a global matching constraint over the learned correspondence structure to exclude cross-view misalignments during the image patch matching process, hence achieving a more reliable matching score between images. Experimental results on various datasets demonstrate the effectiveness of our approach.
@article{arxiv.1504.06243,
title = {Person Re-identification with Correspondence Structure Learning},
author = {Yang Shen and Weiyao Lin and Junchi Yan and Mingliang Xu and Jianxin Wu and Jingdong Wang},
journal= {arXiv preprint arXiv:1504.06243},
year = {2016}
}