Awesome
Perceptual IQA Dataset (PIPAL) and Codebase
PIPAL: a Large-Scale Image Quality Assessment Dataset for Perceptual Image Restoration
<a href="https://www.jasongt.com" target="_blank">Jinjin Gu</a>, <a href="http://www.haomingcai.com" target="_blank">Haoming Cai</a>, <a href="https://chenhaoyu.com" target="_blank">Haoyu Chen</a>, <a href="https://www.linkedin.com/in/yexiaoxing/" target="_blank">Haoyu Chen</a>, <a href="http://www.jimmyren.com" target="_blank">Jimmy S.Ren</a>, <a href="http://xpixel.group/2010/01/20/chaodong.html" target="_blank">Chao Dong</a>. In ECCV, 2020.
- ECCV 2020 Paper | Journal Version | Project Web | NTIRE 2021 Challenge.
- If you have any questions, please contact with haomingcai@link.cuhk.edu.cn
🔥 Important Notes [ 2022-Jan-02 ]
-
[2022-Jan-02] Update the README. I will recorrect those errors mentioned in the issues recently. Moreover, we will extend the PIPAL dataset with more degradations for the NTIRE 2022 (We are still waiting for further notes from the NTIRE organizer).
-
[2021-Feb] We are organizing NTIRE 2021 Perceptual IQA Challenge !!.
-
[2021-Feb] ❗️ ❗️ This codebase ONLY supports users to train LPIPS on PIPAL or BAPPS for now. The SWD module will be added in the future.
📦 Download PIPAL NTIRE 2021
- Train [Google Drive]
- Valid [Google Drive]
- Test [Coming Soon]
🔧 Dependencies and Installation
- Python 3 (Recommend to use Anaconda)
- PyTorch >= 1.0
- NVIDIA GPU + CUDA
- Python packages:
pip install numpy opencv-python lmdb pyyaml - TensorBoard:
- PyTorch >= 1.1:
pip install tb-nightly future - PyTorch == 1.0:
pip install tensorboardX
- PyTorch >= 1.1:
💻 How to Train
- Train Your IQA
- Prepare IQA dataset PIPAL public training [NTIRE 2021] or BAPPS. More details are in
codes/data. - Modify the dataset format based on your need in
codes/data/PairedTrain_dataset.pyandValidorTest_dataset.py - Modify the configuration file
codes/options/train_test_yml/train_our_IQA.yml - Run command:
python train.py -opt options/train_test_yml/train_IQA.yml - Prepare IQA dataset PIPAL public training [NTIRE 2021] or BAPPS. More details are in
📈 How to Test
- Test your IQA
- Prepare IQA dataset PIPAL public validation [NTIRE 2021]. More details are in
codes/data. - Modify the dataset format based on your need in
ValidorTest_dataset.py - Modify the configuration file
codes/options/train_test_yml/test_IQA.yml - Run command:
python test.py -opt options/train_test_yml/test_IQA.yml - Prepare IQA dataset PIPAL public validation [NTIRE 2021]. More details are in
Ackowledgement
- This code is based on mmsr.