Update README.md

README.md

@@ -24,6 +24,6 @@ The model is trained on a subset (10,000 images) of [Crowd Instance-level Human
 The model uses ResNet50 pretrained on ImageNet as the backbone model.
 
 References:   
-[Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://arxiv.org/pdf/1802.02611.pdf)   
-[Rethinking Atrous Convolution for Semantic Image Segmentation](https://arxiv.org/abs/1706.05587)   
-[DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://arxiv.org/abs/1606.00915)
+1. [Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://arxiv.org/pdf/1802.02611.pdf)   
+2. [Rethinking Atrous Convolution for Semantic Image Segmentation](https://arxiv.org/abs/1706.05587)   
+3. [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://arxiv.org/abs/1606.00915)