Code of the paper "DVSRNet: Deep Video Super-Resolution Based on Progressive Deformable Alignment and Temporal-Sparse Enhancement".
CUDA==11.6 Python==3.7 Pytorch==1.13
conda create -n DVSRNet python=3.7 -y && conda activate DVSRNet
git clone --depth=1 https://github.com/QZ1-boy/DVSRNet && cd QZ1-boy/DVSRNet/
# given CUDA 11.6
python -m pip install torch==1.13.1+cu116 torchvision==0.14.1+cu116 torchaudio==0.13.1 --extra-index-url https://download.pytorch.org/whl/cu116
python -m pip install tqdm lmdb pyyaml opencv-python scikit-imageVSR Training Dataset:
[MMCNN dataset] MMCNN
VSR Testing Datasets:
[Myanmar dataset] Myanmar, [YUV21 dataset] YUV21, [Vid4 dataset] Vid4
Optical Flow Training and Testing Datasets:
[Sintel dataset] Sintel, [KITTI dataset] KITTI, [FlyingChairs dataset] FlyChairs, [FlyingThings3D dataset] FlyThings3D
Models： [Pre-trained Models] Pre-trained Models[Code][DVSR]
python basicsr/train.py -opt options/train/ICME2/train_VSRFG_TSEM_spynet_S_MM522_VMY_x4.ymlpython basicsr/test.py -opt options/test/ICME2/test_VSRFG_TSEM_spynet_S_MM522_VMY_x4.ymlIf this repository is helpful to your research, please cite our paper:
@article{zhu2024dvsrnet,
title={Dvsrnet: Deep video super-resolution based on progressive deformable alignment and temporal-sparse enhancement},
author={Zhu, Qiang and Chen, Feiyu and Zhu, Shuyuan and Liu, Yu and Zhou, Xue and Xiong, Ruiqin and Zeng, Bing},
journal={IEEE Transactions on Neural Networks and Learning Systems},
year={2024},
publisher={IEEE}
}