基于视频的行人重识别时空互促

# Spatial and Temporal Mutual Promotion for Video-based Person Re-identification ### News **2018.12.29** The complete code is already available in [person-reid-lib] **2018.11.19** The code of a Refining Recurrent Unit (RRU) has been uploaded. We are integrating our code with [deep-reid] The new code library will includes image-based re-id and video-based re-id. The optical flow and web page visualization are also supported. It will be released in [person-reid-lib] ### Tips for the implementation of our methods in Spatial and Temporal Mutual Promotion for Video-based Person Re-identification * Enough GPU: We use 4 1080ti. * Enough trainign epoches: you can see the differences by testing from 100 epoches to 900 epoches for iLIDS and PRID-2011.