Hitchhiker's Guide to Super-Resolution: Introduction and Recent Advances

TL;DR

This paper reviews recent advances in deep learning-based super-resolution, highlighting new models, techniques, and evaluation methods to guide future research in the field.

Contribution

It provides a comprehensive overview of state-of-the-art SR models, incorporating latest developments like uncertainty-driven losses, wavelet networks, and neural architecture search.

Findings

Diffusion and transformer-based SR models show promising results.

Emerging techniques improve flexibility and evaluation in SR.

The review identifies unexplored research directions.

Abstract

With the advent of Deep Learning (DL), Super-Resolution (SR) has also become a thriving research area. However, despite promising results, the field still faces challenges that require further research e.g., allowing flexible upsampling, more effective loss functions, and better evaluation metrics. We review the domain of SR in light of recent advances, and examine state-of-the-art models such as diffusion (DDPM) and transformer-based SR models. We present a critical discussion on contemporary strategies used in SR, and identify promising yet unexplored research directions. We complement previous surveys by incorporating the latest developments in the field such as uncertainty-driven losses, wavelet networks, neural architecture search, novel normalization methods, and the latests evaluation techniques. We also include several visualizations for the models and methods throughout each chapter in order to facilitate a global understanding of the trends in the field. This review is ultimately aimed at helping researchers to push the boundaries of DL applied to SR.