Arbitrary-scale Super-resolution via Deep Learning: A Comprehensive Survey

Hongying Liu, Zekun Li, Fanhua Shang, Yuanyuan Liu, Liang Wang, Wei Feng, Radu Timofte

Abstract: Highlights•This work is the first systematic review on arbitrary scale super-resolution (SR).•Two novel taxonomies for arbitrary scale SR methods are proposed.•The advantages and limitations of each class of methods are analyzed.•The evaluation on the performance of 24 algorithms are shown.•The trends and future directions are discussed.