Robust Single-Image Super-Resolution via CNNs and TV-TV Minimization

Marija Vella, João F. C. Mota

Research output: Contribution to journalArticlepeer-review

Abstract

Single-image super-resolution is the process of increasing the resolution of an image, obtaining a high-resolution (HR) image from a low-resolution (LR) one. By leveraging large training datasets, convolutional neural networks (CNNs) currently achieve the state-of-the-art performance in this task. Yet, during testing/deployment, they fail to enforce consistency between the HR and LR images: if we downsample the output HR image, it never matches its LR input. Based on this observation, we propose to post-process the CNN outputs with an optimization problem that we call TV-TV minimization, which enforces consistency. As our extensive experiments show, such post-processing not only improves the quality of the images, in terms of PSNR and SSIM, but also makes the super-resolution task robust to operator mismatch, i.e., when the true downsampling operator is different from the one used to create the training dataset.

Original languageEnglish
JournalIEEE Transactions on Image Processing
DOIs
Publication statusAccepted/In press - 16 Aug 2021

Keywords

  • ℓ1-ℓ1 minimization
  • Image reconstruction
  • Image super-resolution
  • Minimization
  • Optimization
  • TV
  • Task analysis
  • Testing
  • Training
  • convolutional neural networks (CNNs)
  • image reconstruction
  • prior information

ASJC Scopus subject areas

  • Software
  • Computer Graphics and Computer-Aided Design

Fingerprint

Dive into the research topics of 'Robust Single-Image Super-Resolution via CNNs and TV-TV Minimization'. Together they form a unique fingerprint.

Cite this