STAN

PyTorch code for our TIP 2022 paper “Video Super-Resolution via a Spatio-Temporal Alignment Network” Paper or Researchgate.

1. Overview
2. Train
3. Test
4. Citation
5. Acknowledgements

Overview

Train

Prerequisites

CUDA 10.2/gcc 5.4/PyTorch 1.4

Train a new model:

bash install.sh
python train.py -datasets_tasks W3_D1_C1_I1

Test

lists/train_tasks_W3_D1_C1_I1.txt specifies the dataset-task pairs for training and testing.

Test a model:

python test.py -method model_name -epoch N -dataset REDS4 -task SR_color/super-resolution

Pretrained Models

The pretrained model for STAN can be found from here.

Citation

If you find the code and paper useful in your research, please cite:

@article{wen2022video,
  title={Video Super-Resolution via a Spatio-Temporal Alignment Network},
  author={Wen, Weilei and Ren, Wenqi and Shi, Yinghuan and Nie, Yunfeng and Zhang, Jingang and Cao, Xiaochun},
  journal={IEEE Transactions on Image Processing},
  volume={31},
  pages={1761--1773},
  year={2022},
  publisher={IEEE}
}

Acknowledgement

This project is based on [Learning Blind Video Temporal Consistency] and our filter adaptive alignment module is based on[STFAN].

@inproceedings{zhou2019stfan,
  title={Spatio-Temporal Filter Adaptive Network for Video Deblurring},
  author={Zhou, Shangchen and Zhang, Jiawei and Pan, Jinshan and Xie, Haozhe and  Zuo, Wangmeng and Ren, Jimmy},
  booktitle={Proceedings of the IEEE International Conference on Computer Vision},
  year={2019}
}

@inproceedings{Lai-ECCV-2018,
    author    = {Lai, Wei-Sheng and Huang, Jia-Bin and Wang, Oliver and Shechtman, Eli and Yumer, Ersin and Yang, Ming-Hsuan}, 
    title     = {Learning Blind Video Temporal Consistency}, 
    booktitle = {European Conference on Computer Vision},
    year      = {2018}
}