Accurate Image Super-Resolution Using Very Deep Convolutional Networks

Code

• github.com/SJHNJU/VDSR
  tf★ 0
• github.com/Lornatang/VDSR-PyTorch
  pytorch★ 0
• github.com/Lornatang/DRRN-PyTorch
  pytorch★ 0
• github.com/GuillaumeDufau/Image-super-resolution-vdsr
  tf★ 0
• github.com/yslenjoy/machine_learning_SuperResolution
  none★ 0
• github.com/jshermeyer/VDSR4Geo
  tf★ 0
• github.com/murrman95/INF573Project2018
  tf★ 0
• github.com/Nhat-Thanh/VDSR-Pytorch
  pytorch★ 0

Abstract

We present a highly accurate single-image super-resolution (SR) method. Our method uses a very deep convolutional network inspired by VGG-net used for ImageNet classification simonyan2015very. We find increasing our network depth shows a significant improvement in accuracy. Our final model uses 20 weight layers. By cascading small filters many times in a deep network structure, contextual information over large image regions is exploited in an efficient way. With very deep networks, however, convergence speed becomes a critical issue during training. We propose a simple yet effective training procedure. We learn residuals only and use extremely high learning rates (10^4 times higher than SRCNN dong2015image) enabled by adjustable gradient clipping. Our proposed method performs better than existing methods in accuracy and visual improvements in our results are easily noticeable.

Tasks

Benchmark Results

Reproductions