An attempt at underwater image lightweight super-resolution using transformer and frequency-domain learning

Lightweight image Super-resolution (SR) is a computer vision technology that aims to recover high-quality image details from low-resolution images with limited computing costs. While Transformer-based SR models have made remarkable advancements, their balanced edge-end deployment and reconstruction...

Popoln opis

Bibliografske podrobnosti

Principais autores:	Haoran Wu, Haitao Wang, Yifan Zhang, Muhammad Yasir Faheem, Xinghua Wang
Format:	Artigo
Jezik:	Inglês
Izdano:	Frontiers Media S.A. 2026-05-01
Serija:	Frontiers in Marine Science
Teme:	deep learning frequency learning lightweight image super-resolution single-image super-resolution underwater image
Online dostop:	https://www.frontiersin.org/articles/10.3389/fmars.2026.1762772/full
Oznake:	Označite Brez oznak, prvi označite!