Deep Back-Projection Networks For Super-Resolution

Code

• github.com/2023-MindSpore-1/ms-code-210/tree/main/DBPN
  mindspore★ 0
• github.com/rkem1542/EDSR-pytorch
  pytorch★ 0
• github.com/tinnunculus/Denoising_SR
  tf★ 0
• github.com/SimoneDutto/EDSR
  pytorch★ 0
• github.com/zhusiling/EDSR
  pytorch★ 0
• github.com/LWChen20/RCAN
  pytorch★ 0
• github.com/akashpalrecha/superres-deformable
  pytorch★ 0
• github.com/laowng/GISR
  pytorch★ 0
• github.com/rajatthepagal/DBPN-Keras
  tf★ 0
• github.com/sanghyun-son/EDSR-PyTorch
  pytorch★ 0

Abstract

The feed-forward architectures of recently proposed deep super-resolution networks learn representations of low-resolution inputs, and the non-linear mapping from those to high-resolution output. However, this approach does not fully address the mutual dependencies of low- and high-resolution images. We propose Deep Back-Projection Networks (DBPN), that exploit iterative up- and down-sampling layers, providing an error feedback mechanism for projection errors at each stage. We construct mutually-connected up- and down-sampling stages each of which represents different types of image degradation and high-resolution components. We show that extending this idea to allow concatenation of features across up- and down-sampling stages (Dense DBPN) allows us to reconstruct further improve super-resolution, yielding superior results and in particular establishing new state of the art results for large scaling factors such as 8x across multiple data sets.

Tasks

Benchmark Results

Reproductions