UVENet: A novel end-to-end model for temporal consistency in underwater video enhancement

Open Webpage

Huijie Guo, Dazhao Du, Hongwei Dong, Shouyou Huang, Changwen Zheng, Lingyu Si

Published: 01 Jan 2026, Last Modified: 04 Nov 2025Neural Networks 2026EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Introduced SUVE, the first synthetic dataset for underwater video enhancement.•Proposed UVENet, an end-to-end model resolving temporal consistency in UVE.•Extensive experiments show UVENet outperforms state-of-the-art UIE methods.