add masked images

youweiliang · May 5, 2022 · 29c7f2a · 29c7f2a
1 parent 5a3fdbd
commit 29c7f2a
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 1 deletion.
diff --git a/README.md b/README.md
@@ -5,7 +5,11 @@ This repository contains PyTorch evaluation code, training code and pretrained _
 > 
 > _Youwei Liang, Chongjian Ge, Zhan Tong, Yibing Song, Jue Wang, Pengtao Xie_
 
-The proposed EViT models obtain competitive tradeoffs in terms of speed / precision:
+The proposed EViT method reduces computation by progressively discarding or fusing inattentive tokens in Vision Transformers, as illustrated in the following figure.
+
+![IMG](images/masked_img.png)
+
+EViT obtains competitive tradeoffs in terms of speed / precision:
 
 ![EViT](images/tradeoff.png)
 

diff --git a/images/masked_img.png b/images/masked_img.png