English

Self-supervised Fine-tuning for Correcting Super-Resolution Convolutional Neural Networks

Image and Video Processing 2020-06-16 v3 Machine Learning Machine Learning

Abstract

While Convolutional Neural Networks (CNNs) trained for image and video super-resolution (SR) regularly achieve new state-of-the-art performance, they also suffer from significant drawbacks. One of their limitations is their lack of robustness to unseen image formation models during training. Other limitations include the generation of artifacts and hallucinated content when training Generative Adversarial Networks (GANs) for SR. While the Deep Learning literature focuses on presenting new training schemes and settings to resolve these various issues, we show that one can avoid training and correct for SR results with a fully self-supervised fine-tuning approach. More specifically, at test time, given an image and its known image formation model, we fine-tune the parameters of the trained network and iteratively update them using a data fidelity loss. We apply our fine-tuning algorithm on multiple image and video SR CNNs and show that it can successfully correct for a sub-optimal SR solution by entirely relying on internal learning at test time. We apply our method on the problem of fine-tuning for unseen image formation models and on removal of artifacts introduced by GANs.

Keywords

Cite

@article{arxiv.1912.12879,
  title  = {Self-supervised Fine-tuning for Correcting Super-Resolution Convolutional Neural Networks},
  author = {Alice Lucas and Santiago Lopez-Tapia and Rafael Molina and Aggelos K. Katsaggelos},
  journal= {arXiv preprint arXiv:1912.12879},
  year   = {2020}
}

Comments

15 pages, 11 figures

R2 v1 2026-06-23T12:58:52.055Z