Jointly Temporal Pooling Networks and Multi-loss Fusion for Video-based Person Re-Identification

with the increasing demand in surveillance and camera networks, video-based person re-identification is an important task in the field of computer vision. The video contains rich samples of person's appearances, how to effectively use this information is a challenge. We propose a hierarchical network that joint temporal pooling and multi-loss fusion function to obtain the representation of videos. The network can be trained efficiently with end-to-end way. We conducted experiments on the large-scale MARS, iLIDS-VID and PRID-2011 datasets to confirm the effectiveness of our proposed method.