Semantic-Guided Pixel Sampling for Cloth-Changing Person Re-Identification

Cloth-changing person re-identification (re-ID) is a new rising research topic that aims at retrieving pedestrians whose clothes are changed. This task quite challenging and has not been fully studied to date. Current works mainly focus on body shape or contour sketch, but they robust enough due view posture variations. The key this exploit cloth-irrelevant cues. paper proposes semantic-guided pixel sampling approach for the cloth-changing re-ID task. We do explicitly define which feature extract force model automatically learn Specifically, we first recognize pedestrian's upper pants, then randomly change them by pixels from other pedestrians. changed samples retain identity labels exchange of pants among different Besides, adopt loss function constrain learned features keep consistent before after changes. In way, forced cues irrelevant pants. conduct extensive experiments latest released PRCC dataset. Our method achieved 65.8% Rank1 accuracy, outperforms previous methods with large margin. code available https://github.com/shuxjweb/pixel_sampling.git.