update introduction

README.md

@@ -1,3 +1,32 @@
 # sketch2normal
 
-This is the implementation of the paper "Interactive Sketch-Based Normal Map Generation with Deep Neural Networks".
+This is the implementation of the paper [Interactive Sketch-Based Normal Map Generation with Deep Neural Networks](http://sweb.cityu.edu.hk/hongbofu/doc/sketch2normal_i3D2018.pdf)
+
+<img src="./teaser.png" width="600px"/>
+
+## Setup
+
+We run the program on a Linux desktop using python.
+
+## Usage
+
+- Train the model:
+'''bash
+pyhton main.py --phase train --dataset_name <dataset>
+'''
+
+- Test:
+'''bash
+python main.py --phase test --dataset_name <dataset>
+'''
+
+- Visulize the training process:
+'''bash
+cd logs
+tensorboard --logdir=./
+'''
+
+- Data: TBA
+
+## Acknowledgement
+This code is based on the implementation of pix2pix from [pix2pix-tensorflow](https://github.com/yenchenlin/pix2pix-tensorflow). Thanks for the great work!