Five A$^{+}$ Network: You Only Need 9K Parameters for Underwater Image Enhancement

TL;DR

This work tackles the challenge of real-time underwater image enhancement on resource-constrained platforms by introducing FA$^{+}$Net, an ultra-lightweight network with ~9k parameters. The model adopts a two-stage architecture: a strong prior stage with Multi-Scale Pyramid Module and Multi-branch Color Enhancement Module to address color distortion and detail restoration, and a fine-grained stage reinforced by a Spatial-Frequency Domain Interaction Module that fuses spatial and Fourier-domain information via Fast Fourier Convolution. The authors demonstrate state-of-the-art performance on multiple underwater datasets (UIEB variants) while achieving real-time 1080P processing on high-end GPUs, outperforming heavier prior methods in both quality metrics (PSNR, MSE, UCIQE) and efficiency. Ablation studies validate the effectiveness of downsampling choices and the α parameter in SDFIM, and the work discusses limitations and directions for broader applicability and edge deployment.

Abstract

A lightweight underwater image enhancement network is of great significance for resource-constrained platforms, but balancing model size, computational efficiency, and enhancement performance has proven difficult for previous approaches. In this work, we propose the Five A$^{+}$ Network (FA$^{+}$Net), a highly efficient and lightweight real-time underwater image enhancement network with only $\sim$ 9k parameters and $\sim$ 0.01s processing time. The FA$^{+}$Net employs a two-stage enhancement structure. The strong prior stage aims to decompose challenging underwater degradations into sub-problems, while the fine-grained stage incorporates multi-branch color enhancement module and pixel attention module to amplify the network's perception of details. To the best of our knowledge, FA$^{+}$Net is the only network with the capability of real-time enhancement of 1080P images. Thorough extensive experiments and comprehensive visual comparison, we show that FA$^{+}$Net outperforms previous approaches by obtaining state-of-the-art performance on multiple datasets while significantly reducing both parameter count and computational complexity. The code is open source at https://github.com/Owen718/FiveAPlus-Network.

Figures (3)

• Figure 1: Comparison of recent state-of-the-art methods and our method: We report the computational efficiency (${\rm{\# }}Params$, GFLOPs, and FPS) and numerical scores for two types of restoration quality measurement metrics including PSNR and SSIM, it can be easily observed that our method is remarkably superior to others.
• Figure 2: The overall architecture of Five A$^{+}$ Network: FA$^{+}$Net is composed of a strong prior stage and a fine-grained stage, augmented by the efficient Spatial-frequency Domain Interaction Module. The core components of the network comprise: (c) MPM captures granular details across various scales, endowing the model with potent detail perception; (d) MCEM perform consecutive processing of individual image pixels, thereby enabling our network to achieve precise color restoration; and (e) SDFIM aids the network in sifting valuable feature information from the outputs of diverse components and acquiring global contextual features, and $\alpha$ is a hyperparameter that controls the fusion ratio of spatial-frequency domain information.
• Figure 3: Visual comparison of UIE networks on T90.