xueshengke / WRANSR-keras

This project is the implementation of paper "Xue S. et al., Wavelet-based Residual Attention Network for Image Super-Resolution, Neurocomputing, 2019".

WRANSR-keras

This project is the implementation of paper "Xue S. et al., Wavelet-based Residual Attention Network for Image Super-Resolution, Neurocomputing, 2019".

Environment

• OS: CentOS 7 Linux kernel 3.10.0-514.el7.x86_64
• CPU: Intel Xeon(R) CPU E5-2667 v4 @ 3.20GHz x 32
• Memory: 251.4 GB
• GPU: NVIDIA Tesla P4, 8 GB

Software

• Python 2.7.14
• Tensorflow 1.13
• Keras

Dataset

These datasets are the same as other paper provided. Readers can directly use them or download them from here:

BSDS100, BSDS200, General-100, Set5, Set14, T91, Train_291, Urban100, and DIV2K.

Train

python main.py

Parameters for training

• scale = 2/3/4
• depth = 8
• ratio = 4
• width = 64
• alpha = 0.1
• batch_size = 64
• epochs = 200

Test

python predict.py

Contact

Ph.D. candidate, Shengke Xue

College of Information Science and Electronic Engineering

Zhejiang University, Hangzhou, P.R. China

Email: [e-mail], [e-mail]