Person Re-identification Based on Group Context Strategy

In this paper, we focus on small group context and propose a group context strategy for the person re-identification. We consider the small group context information as a robust cue that used for the non-overlapping surveillance cameras re-identity work. More precisely, we exploit the small group context information by the group context strategy to improve the re-identity work performance. Experimental results demonstrate that group affiliation information is a robust cue and the group context strategy can improve re-identity performance, and our proposed works substantially better than other current state-of-the-art person re-identification methods.