diff --git a/README.md b/README.md
index b9184e8..d1a6e55 100644
--- a/README.md
+++ b/README.md
@@ -1,4 +1,6 @@
 # R\*CNN
+
+[![Join the chat at https://gitter.im/gkioxari/RstarCNN](https://badges.gitter.im/gkioxari/RstarCNN.svg)](https://gitter.im/gkioxari/RstarCNN?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)
 Source code for R\*CNN, created by Georgia Gkioxari at UC Berkeley.
 
 ### Introduction
@@ -92,13 +94,17 @@ Test a R\*CNN classifier
 
 	Download the data from [here](http://www.cs.berkeley.edu/~gkioxari/RstarCNN/BAPD.tar.gz) and place them inside the `$ROOT/data` directory
 
-3. Reference models
+3. Stanford 40 Dataset
+      
+      Download the data from [here](http://www.cs.berkeley.edu/~gkioxari/RstarCNN/Stanford40.tar.gz) and place them inside `$ROOT/data` directory. R*CNN achieves 90.85% on the test set (trained models provided in 5)
+
+4. Reference models
 	
-	Download the VGG16 reference model trained on ImageNet from [here](http://www.cs.berkeley.edu/~gkioxari/RstarCNN/reference_models.tar.gz)
+	Download the VGG16 reference model trained on ImageNet from [here](http://www.cs.berkeley.edu/~gkioxari/RstarCNN/reference_models.tar.gz) (500M)
 
-4. Trained models
+5. Trained models
 	
-	Download the models as described in the paper from [here](http://www.cs.berkeley.edu/~gkioxari/RstarCNN/trained_models.tar.gz)
+	Download the models as described in the paper from [here](http://www.cs.berkeley.edu/~gkioxari/RstarCNN/trained_models.tar.gz) (3.6G)