Update README.md

README.md

@@ -16,7 +16,7 @@ Chunyu Xie*, Bin Wang*, Fanjing Kong, Jincheng Li, Dawei Liang, Gengshen Zhang,
 [![ICML](https://img.shields.io/badge/ICML-2025-blue.svg)](https://icml.cc/Conferences/2025)
 
  <p align="center">
-  <img src="https://github.com/360CVGroup/FG-CLIP/blob/main/use_imgs/radar_chart_methods.png"  width="500" height="440"/>
+  <img src="radar_chart_methods.png"  width="500" height="440"/>
 </p>
 
 ## Model Framework
@@ -24,7 +24,7 @@ FG-CLIP’s training proceeds in two stages: the first stage leverages
 global-level caption-image pairs to achieve initial fine-grained alignment, while the second stage supplements these with additional
 region-level captions, including detailed region captions and positive/negative region descriptions to further refine the alignment.
 <p align="center">
-  <img src="https://github.com/360CVGroup/FG-CLIP/blob/main/use_imgs/fgclip_strc.png" width=80%/>
+  <img src="fgclip_strc.png" width=80%/>
 </p>
 
 ## Quick Start 🤗
@@ -119,7 +119,7 @@ plt.savefig("FG-CLIP/use_imgs/FGCLIP_dfcolor_cat.png")
 ```
 
  <p align="left">
-  <img src="https://github.com/360CVGroup/FG-CLIP/blob/main/use_imgs/FGCLIP_dfcolor_cat.png" width=25%/>
+  <img src="FGCLIP_dfcolor_cat.png" width=25%/>
 </p>
 
 ## Citation