Real-Time Underwater Image Enhancement via Frequency-Guided Dual-Path Attention

2026-06-29

AI Summary

This work addresses the demand for low-latency, compact, and computationally efficient underwater image enhancement suitable for deployment on mobile and robotic platforms. To this end, we propose a lightweight real-time enhancement framework that, for the first time, integrates a fixed DCT frequency-domain prior into a reparameterizable convolutional block (MBRConv-DCT) and introduces a frequency-guided dual-path attention module (FGDPA) to effectively fuse spatial and spectral information. This design overcomes the limitation of existing lightweight methods that neglect frequency-sensitive degradation. With only 4.23K parameters and inference speed exceeding 600 FPS—without incurring additional computational overhead—the model surpasses larger-scale approaches in both quantitative metrics and visual quality, achieving state-of-the-art performance in real-time underwater image enhancement.

Abstract

Real-time underwater image enhancement (UIE) is crucial for mobile underwater photography and autonomous robotic systems, where practical deployment typically requires low latency and compact models under constrained computational resources. Recent ultra-lightweight CNNs based on structural re-parameterization meet these constraints but operate purely in the spatial domain, ignoring the frequency-sensitive nature of underwater degradation. To address this, we propose a lightweight UIE framework that integrates two key components: a Multi-Branch Reparameterizable Convolution with Fixed DCT Priors (MBRConv-DCT) that injects structured directional frequency priors during training, and a Frequency-Guided Dual-Path Attention (FGDPA) module that fuses spatial and spectral cues via a dual-path design for adaptive feature modulation. Both components are fully compatible with structural re-parameterization: the convolution branch introduces zero additional inference cost after re-parameterization, while the attention module incurs only a minimal computational overhead. Experiments show our model achieves state-of-the-art performance with only 4.23K parameters and 600+ FPS, outperforming much larger methods in both quantitative metrics and visual quality. Code is available at https://github.com/LethyZhang/FGDPA.

Problem

• real-time
• underwater image enhancement
• low latency
• compact models
• computational constraints

Innovation

• Frequency-Guided Dual-Path Attention
• Structural Re-parameterization
• DCT Priors
• Lightweight UIE
• Real-Time Underwater Image Enhancement