--- a/README.md
+++ b/README.md
@@ -41,7 +41,7 @@
 
 We use [SegNet](http://mi.eng.cam.ac.uk/projects/segnet/) as one of our base networks to conduct segmentation. We feed our annotated training data and train the network end-to-end, and evaluate it on our annotated testing data. 
 
-![segnet](./doc/segnet.jpg)
+![segnet](https://github.com/kaiwenzha/3D-Teeth-Reconstruction-from-CT-Scans/blob/master/doc/segnet.jpg?raw=true)
 
 ### Dependencies