init

README.md

@@ -1,9 +1,10 @@
-# Statistical Image Processing
+# BLIND USRNET
 
 A project in Statistical Image Processing.
 Extension of [Deep unfolding network for image super-resolution](https://arxiv.org/pdf/2003.10428.pdf) 
 for Blind Super Resolution via [Kernel Estimation](https://arxiv.org/abs/1909.06581) 
 
+This code borrows heavily from [1] & [2], see the papers and code in the Acknowledgment section.
 
 ## Getting Started
 
@@ -15,30 +16,61 @@ git clone https://github.com/geopi1/Improved_USRNet.git
 ### Datasets
 Download the Datasets (a bicubically downscaled version of DIV2K):
 
-[link to mridata](https://drive.google.com/drive/folders/1wITaNr7KNg_keUnK4myKjUoKA4HGeONA?usp=sharing)
+[Download DIV2K](https://drive.google.com/drive/folders/1wITaNr7KNg_keUnK4myKjUoKA4HGeONA?usp=sharing) 
+This link will download 2 folders:
+* DIV2K_LR - the low resolution input
+* DIV2K_HR_ds2 - Ground truth high resolution images
 
 ### Prerequisites
-Tested and run on UBUNTU 18.04
+Tested and run on: 
+ * UBUNTU 18.04 
+ * RTX 2080 
+ * Nvidia driver 440.95.10 
+ * cuda 10.1.243 
+ * cudnn 7.6.5 
+ * pytorch 1.5 
+ * TF 1.14
+
+
 1. Setup conda 
  ```bash
  conda env create -f env.yml
  ```
  This will create a working environment named Blind_USRNet
 2. Setup can also be performed with pip (virtual env) via the requirements.txt file 
  ```bash
- python3 -m venv Blind_USRNet
+ python -m venv Blind_USRNet
  pip install -r requirements.txt
  ```
 3. put images in right place
 
-## Running Tests
+## Testing
 ### Code
-HOW TO RUN
+```bash
+python main.py 
+```
+This will output 4 types of images to ./results: 
+* Default USRNet settings
+* Default USRNet settings + Noise STD estimation
+* Kernel estimated USRNet
+* Kernel estimated USRNet + Noise STD estimation
+
+Additionally, for each estimated image a side-by-side image with the LR version is saved with the degredation kernel (default or estimated) 
+
+### Numerical Evaluation
+To calculate the numeric results on the whole dataset run:
+```bash
+python utils/get_results.py
+```
+This will calculate the mean PSNR and SSIM on the SR<->HR on all the results
 
-### Logs
-RESULTS 
+## Numerical Results
+| Metric | Default | Default + Noise Est. | Kernel Est. | Kernel Est. + Noise Est. |
+| ------------- | ------- | -------------------- | ----------- | ------------------------ |
+| PSNR | | $1600 | | | |
+| SSIM | | $12 | | | |
 
-## Results
+## Visual Results
 ![Default](./results/0002_x4_usrnetdefaultdefault_noise_LE.png)
 
 ![Ours](./results/0002_x4_usrnetKernelGANdefault_noise_LE.png)

main.py

@@ -10,7 +10,7 @@
 
 
 # 'usrgan' | 'usrnet' | 'usrgan_tiny' | 'usrnet_tiny'
-config = {'net': 'usrnet'}
+config = {'net': 'usrgan'}
 # list all images
 ext = ['.png', '.jpg', 'jpeg']