Person Re-identification with Correspondence Structure Learning

2015/04/23 by Shen, Yang, Lin, Weiyao, Yan, Junchi +3
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.1504.06243

Abstract

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.

Related

• Robust and Efficient Graph Correspondence Transfer for Person Re-identification
• Graph Correspondence Transfer for Person Re-identification
• Deeply-Learned Part-Aligned Representations for Person Re-Identification
• Person Re-identification by Saliency Learning
• Hierarchical Invariant Feature Learning with Marginalization for Person Re-Identification