Dual-Branch Network Fused With Two-Level Attention Mechanism for Clothes-Changing Person Re-Identification

Clothes-changing person re-identification is a hot topic in the current academic circles. Most of methods assume that clothes will not change short period time, but they are applicable when people clothes. Based on this situation, paper proposes dual-branch network for clothes-changing integrates two-level attention mechanism and captures aggregates fine-grained semantic information channels spaces through suppresses sensitivity to clothing features by training classification branch. The method does use auxiliary means such as human skeletons, complexity model greatly reduced compared with most methods. This conducts experiments popular dataset PRCC very large-scale cross-spatial-temporal (LaST). experimental results show more advanced than existing