Underwater Image Enhancement via Dehazing and Color Restoration
Chengqin Wu, Shuai Yu, Tuyan Luo, Qiuhua Rao, Qingson Hu, and Jingxiang Xu, Lijun Zhang

TL;DR
This paper introduces WaterFormer, a ViT-based network that separately models haze and color degradation in underwater images, achieving superior enhancement by dynamically fusing these features with physics-based reconstruction.
Contribution
The paper presents a novel ViT-based architecture with decoupled haze and color restoration modules, and a physics-informed reconstruction layer for underwater image enhancement.
Findings
Outperforms state-of-the-art methods in quality metrics
Effectively preserves color fidelity and structural details
Demonstrates robustness across diverse underwater conditions
Abstract
Underwater visual imaging is crucial for marine engineering, but it suffers from low contrast, blurriness, and color degradation, which hinders downstream analysis. Existing underwater image enhancement methods often treat the haze and color cast as a unified degradation process, neglecting their inherent independence while overlooking their synergistic relationship. To overcome this limitation, we propose a Vision Transformer (ViT)-based network (referred to as WaterFormer) to improve underwater image quality. WaterFormer contains three major components: a dehazing block (DehazeFormer Block) to capture the self-correlated haze features and extract deep-level features, a Color Restoration Block (CRB) to capture self-correlated color cast features, and a Channel Fusion Block (CFB) that dynamically integrates these decoupled features to achieve comprehensive enhancement. To ensure…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsImage Enhancement Techniques · Image and Signal Denoising Methods · Advanced Image Fusion Techniques
MethodsAttention Is All You Need · Byte Pair Encoding · Absolute Position Encodings · Vision Transformer · Softmax · Label Smoothing · Layer Normalization · Dropout · Position-Wise Feed-Forward Layer · Residual Connection
