Video Super Resolution

A collection of state-of-the-art video or single-image super-resolution architectures, reimplemented in tensorflow. (Now VEPSCN model has finally been added to Models, pretrained weights will be uploaded soon...)

Network list and reference

The hyperlink directs to paper site, follows the official codes if the authors open sources.

Classic
1. SRCNN
2. RAISR
CVPR 2016
1. Efficient Sub-Pixel Convolutional Network: ESPCN
2. Very Deep Convolutional Networks: VDSR
3. Deeply-Recursive Convolutional Network: DRCN
CVPR 2017
1. Deep Recursive Residual Network: DRRN code
2. Deep Laplacian Pyramid Networks: LapSRN code
3. Enhanced Deep Residual Networks: EDSR
ICCV 2017
1. Memory Network: MemNet code
CVPR 2018
1. Information Distillation Network: IDN code
2. Residual Dense Network: RDN code
3. Super-Resolution Network for Multiple Degradations: SRMD code
4. Deep Back-Projection Networks: DBPN code
5. Zero-Shot Super-Resolution: ZSSR code
ECCV 2018
1. Cascading Residual Network: CARN code
2. Residual Channel Attention Networks: RCAN code
Others
1. DNCNN (This is for denoise) code
2. Deep CNN with Skip Connection: DCSCN code
Videos
1. VESPCN
2. SPMC code
3. FRVSR code
4. DUF code

All these models are implemented in ONE framework.

Link of datasets

(please contact me if any of links offend you or any one disabled)

[THANKS] Above links are from jbhuang0604

VSR package

This package offers a training and data processing framework based on TF. What I made is a simple, easy-to-use framework without lots of encapulations and abstractions. Moreover, VSR can handle raw NV12/YUV as well as a sequence of images as inputs.

Install

git clone https://github.com/loseall/VideoSuperResolution && cd VideoSuperResolution
pip install -e .

How to use

To train/test/infer any model in VSR.Models, please see README. To write and train your own model via VSR, please see Docs.

Todo

FRVSR
ZSSR

caoyangcr7 / VideoSuperResolution