Semantic Part Constraint for Person Re-identification

In order to alleviate the background clutter in pedestrian images, and make the network focus on pedestrian foreground to improve the utilization of human body parts in the foreground. In this paper, a person re-identification network is proposed that introduces Semantic Part Constraint(SPC). Firstly, the pedestrian image is input into the backbone network and the semantic part segmentation network at the same time, and the pedestrian feature map and the part segmentation label are obtained respectively. Secondly, the part segmentation label and the pedestrian feature maps are merged to obtain the semantic part feature. Thirdly, the pedestrian feature map is obtained and the global average pooling is used to gain global features. Finally, the network is trained using both identity constraint and semantic part constraint. Since the semantic part constraint makes the global features obtain the part information, only the backbone network can be used to extract the features of the pedestrian during the test. Experiments on large-scale datasets show that semantic part constraints can effectively make the network improve the ability to identify pedestrians and reduce the computational cost of inferring networks. Compared with the state of art, the proposed network can better resist background clutter and improve person re-identification performance.