diff --git a/README.md b/README.md
index 49f00fe..9b15da0 100644
--- a/README.md
+++ b/README.md
@@ -31,6 +31,9 @@ The neighbor encoding method of our paper is inspired by the reference [2]. If y
 1. re_ranking_feature.py: re-ranking with original feature, Euclidean distance is used. Thanks [Hao Luo](https://github.com/michuanhaohao) !
 2. re_ranking_ranklist: re-ranking with given distance matrices, handle the difference of / division between python 2 and 3. Thanks [huang houjing](https://github.com/huanghoujing)  !
 
+### Pytorch re-implementations
+-[Person_reID_baseline + Random Erasing + Re-ranking](https://github.com/layumi/Person_reID_baseline_pytorch)
+
 ================================================================ 
 
 ## The new training/testing protocol for CUHK03