Update README.md

Author: chenliu-wustl

README.md

@@ -5,11 +5,17 @@ Python 2.7, TensorFlow (>= 1.0), numpy, opencv 3.
 
 ## Getting started
 ### Compilation
-Please run the following command to compile the library for computing the set matching loss. (See [here](https://github.com/fanhqme/PointSetGeneration) for details.)
+Please run the *make* command under the root folder to compile the library for computing the set matching loss. (See [here](https://github.com/fanhqme/PointSetGeneration) for details.)
 
+### Data preparation
+We convert [ScanNet](http://www.scan-net.org/) data to *.tfrecords* files for training and testing. The *.tfrecords* file can be downloaded from [here](https://mega.nz/#!IvAixABb!PD3wJtXX_6W3qtfKZQtl_P07mYPLwWst3cwbvuTXlSY).
 
+### Train the network
 To train the network from the pretrained DeepLab network, please first download the DeepLab model [here](https://github.com/DrSleep/tensorflow-deeplab-resnet) (under the Caffe to TensorFlow conversion), and then run the following command.
 
 ```bash
-python train_planenet.py --restore=0
+python train_planenet.py --restore=0 --modelPathDeepLab="path to the deep lab model" --rootFolder="folder which contains tfrecords files"
 ```
+
+### Evaluation
+Please first download our trained network from [here](https://mega.nz/#!sjpT2DiQ!Uo-6hxyldmtnPoKk3TTdUHKZADRGy6nIPlmAeVzJs_8) and put the uncompressed folder under ./checkpoint