A Review on Self Learning Based Methods for Real World Single Image Super Resolution

EasyChair Preprint 6721

Abstract

Image super resolution (ISR) is one of the popular techniques of image processing to boost the resolution of images. Reconstructing high-resolution (HR) image from low resolution (LR) degraded images results in Single image super resolution (RSISR) reconstruction. In the domain of image processing, it is the lively research topic. This paper covers datasets which are available and assessment metrics for RSISR and method of RSISR based on Self-Learning RSISR [1]. In terms of both reconstruction quality and computational efficiency comparisons are done among representative RSISR methods on datasets. We will discuss challenges on RSISR.

Keyphrases: Real-world image, Self-learning-based methods, computer vision, datasets, deep learning, super-resolution

@booklet{EasyChair:6721,
  author    = {Yogesh Gaikwad},
  title     = {A Review on Self Learning Based Methods for Real World Single Image Super Resolution},
  howpublished = {EasyChair Preprint 6721},
  year      = {EasyChair, 2021}}