RQUL-UIE: Revitalizing Quality-Unstable Labels for Underwater Image Enhancement via In-Dataset Self-Supervision

2026-06-04

AI Summary

This work addresses the limitations of existing underwater image enhancement methods, which rely on paired training data with inconsistent label quality that hinders model performance. The authors propose a novel training-free self-supervised strategy that leverages semantic-aware embeddings from a pre-trained diffusion model to assess label quality, quantifying it as a noise-level index to guide multi-step denoising with hierarchical supervision. Additionally, a Fourier-domain refinement network is introduced to recover high-frequency details. This approach uniquely integrates training-free quality assessment with hierarchical self-supervision in underwater image enhancement, effectively utilizing low-quality labels without performance degradation. Experimental results demonstrate that the method significantly outperforms current state-of-the-art techniques across multiple evaluation metrics.

Abstract

Underwater Image Enhancement (UIE) is essential for mitigating degradations caused by water medium. Although learning-based methods have advanced significantly, most rely on paired datasets with unstable label quality, which bottlenecks model performance. This paper proposes a diffusion-based, in-dataset self-supervised learning strategy designed to exploit the quality distribution of training labels. Specifically, we evaluate label quality via semantic perception embeddings from a pre-trained diffusion model in a training-free manner. These quality scores are subsequently quantized into noise-level indices, guiding a multi-step denoising process for level-wise supervision. This mechanism prevents low-quality labels from degrading the model while maximizing their utility during training. Furthermore, a Fourier-based refinement network is incorporated to explicitly reconstruct high-frequency components. Extensive evaluations demonstrate that our method consistently outperforms SOTA approaches in restoration quality. The code and pre-trained model will be available once accepted in link.

Problem

• Underwater Image Enhancement
• Quality-Unstable Labels
• Label Quality
• Paired Datasets
• Model Performance Bottleneck

Innovation

• self-supervised learning
• diffusion model
• label quality assessment
• multi-step denoising
• Fourier-based refinement