Update README.md

README.md

@@ -21,14 +21,12 @@ and [Ming-Hsuan Yang](http://faculty.ucmerced.edu/mhyang/)
 1. [Dataset](#dataset)
 1. [Apply Pre-trained Models](#apply-pre-trained-models)
 1. [Training and Testing](#training-and-testing)
-1. [Image processing algorithms](#image-processing-algorithms)
+1. [Image Processing Algorithms](#image-processing-algorithms)
 
 ### Introduction
 Our method takes the original unprocessed and per-frame processed videos as inputs to produce a temporally consistent video. Our approach is agnostic to specific image processing algorithms applied on the original video.
 
 
-
-
 ### Citation
 
 If you find the code and datasets useful in your research, please cite:
@@ -44,25 +42,64 @@ If you find the code and datasets useful in your research, please cite:
 ### Requirements and dependencies
 - [Pytorch 0.4](https://pytorch.org/)
 - [TensorboardX](https://github.com/lanpa/tensorboardX)
+- [FlowNet2-Pytorch](https://github.com/NVIDIA/flownet2-pytorch) (Code and model are already included in this repository)
 - [LPIPS](https://github.com/richzhang/PerceptualSimilarity) (for evaluation)
 
-
-### Dataset
-
+Our code is tested on Ubuntu 16.04 with cuda 9.0 and cudnn 7.0.
 
 ### Installation
 Download repository:
 
-    $ git clone https://github.com/phoenix104104/fast_blind_video_consistency.git
+    git clone https://github.com/phoenix104104/fast_blind_video_consistency.git
+
+Compile FlowNet2 dependencies (correlation, resample, and channel norm layers):
+
+    ./install.sh
+
+### Dataset
+Download our training and testing datasets:
+
+    cd data
+    ./download_data.sh [train | test | all]
+
+For example, download training data only:
 
+    ./download_data.sh train
+
+Download both training and testing data:
 
+    ./download_data.sh all
+
+You can also download the results of [Bonneel et al. 2015] and our approach:
+
+    ./download_data.sh results
+
 ### Apply pre-trained models
+Download pretrained models (including FlowNet2 and our model):
+
+    cd pretrained_models
+    ./download_models.sh
 
 
 ### Training and testing
 
-### Image processing algorithms
+
+### Image Processing Algorithms
 We use the following algorithms to obtain per-frame processed results:
 
 **Style transfer**
+- [WCT: Universal Style Transfer via Feature Transforms, NIPS 2017](https://github.com/Yijunmaverick/UniversalStyleTransfer)
+- [Fast Neural Style Transfer: Perceptual Losses for Real-Time Style Transfer and Super-Resolution, ECCV 2016](https://github.com/jcjohnson/fast-neural-style)
+
+**Image Enhancement**
+- [DBL: Deep Bilateral Learning for Real-Time Image Enhancement, Siggraph 2017](https://groups.csail.mit.edu/graphics/hdrnet/)
+
+**Intrinsic Image Decomposition**
+- [Intrinsic Images in the Wild, Siggraph 2014](http://opensurfaces.cs.cornell.edu/publications/intrinsic/)
+
+**Image-to-Image Translation**
+- [CycleGAN: Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks, ICCV 2017](https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix)
 
+**Colorization**
+- [Colorful Image Colorization, ECCV 2016](https://github.com/richzhang/colorization)
+- [Let there be Color!: Joint End-to-end Learning of Global and Local Image Priors for Automatic Image Colorization with Simultaneous Classification, Siggraph 2016](http://hi.cs.waseda.ac.jp/~iizuka/projects/colorization/en/)