Style-Decoupled Adaptive Routing Network for Underwater Image Enhancement

2026-04-14

AI Summary

Existing underwater image enhancement methods rely on uniform mapping strategies, which struggle to accommodate the diverse restoration requirements of images suffering from varying degrees of degradation. To address this limitation, this work proposes SDAR-Net, a novel framework that explicitly decouples the input image into degradation style and scene structure for the first time. It introduces a style-embedding-based adaptive soft-weight routing mechanism to dynamically fuse multi-state enhancement representations, enabling precise and adaptive restoration across a spectrum from mildly to severely degraded underwater images. Evaluated on real-world underwater image benchmarks, the proposed method achieves a state-of-the-art PSNR of 25.72 dB and significantly improves performance in downstream vision tasks.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Intelligent Robots: State Estimation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Underwater Image Enhancement (UIE) is essential for robust visual perception in marine applications. However, existing methods predominantly rely on uniform mapping tailored to average dataset distributions, leading to over-processing mildly degraded images or insufficient recovery for severe ones. To address this challenge, we propose a novel adaptive enhancement framework, SDAR-Net. Unlike existing uniform paradigms, it first decouples specific degradation styles from the input and subsequently modulates the enhancement process adaptively. Specifically, since underwater degradation primarily shifts the appearance while keeping the scene structure, SDAR-Net formulates image features into dynamic degradation style embeddings and static scene structural representations through a carefully designed training framework. Subsequently, we introduce an adaptive routing mechanism. By evaluating style features and adaptively predicting soft weights at different enhancement states, it guides the weighted fusion of the corresponding image representations, accurately satisfying the adaptive restoration demands of each image. Extensive experiments show that SDAR-Net achieves a new state-of-the-art (SOTA) performance with a PSNR of 25.72 dB on real-world benchmark, and demonstrates its utility in downstream vision tasks. Our code is available at https://github.com/WHU-USI3DV/SDAR-Net.

Problem

• Underwater Image Enhancement
• degradation styles
• adaptive restoration
• uniform mapping
• image over-processing

Innovation

• style decoupling
• adaptive routing
• underwater image enhancement
• degradation-aware enhancement
• feature modulation