SFormer: SNR-guided Transformer for Underwater Image Enhancement from the Frequency Domain

TL;DR

SFormer introduces a frequency domain SNR-guided transformer architecture for underwater image enhancement, effectively separating interference and amplifying important features to produce clearer, more detailed underwater images.

Contribution

The paper proposes a novel frequency domain SNR prior and a transformer-based architecture, SFormer, that improves underwater image enhancement by better channel modulation and spectral interaction.

Findings

Achieves 3.1 dB higher PSNR than recent methods

Improves SSIM by 0.08 over existing approaches

Restores colors, textures, and contrast effectively

Abstract

Recent learning-based underwater image enhancement (UIE) methods have advanced by incorporating physical priors into deep neural networks, particularly using the signal-to-noise ratio (SNR) prior to reduce wavelength-dependent attenuation. However, spatial domain SNR priors have two limitations: (i) they cannot effectively separate cross-channel interference, and (ii) they provide limited help in amplifying informative structures while suppressing noise. To overcome these, we propose using the SNR prior in the frequency domain, decomposing features into amplitude and phase spectra for better channel modulation. We introduce the Fourier Attention SNR-prior Transformer (FAST), combining spectral interactions with SNR cues to highlight key spectral components. Additionally, the Frequency Adaptive Transformer (FAT) bottleneck merges low- and high-frequency branches using a gated attention mechanism to enhance perceptual quality. Embedded in a unified U-shaped architecture, these modules integrate a conventional RGB stream with an SNR-guided branch, forming SFormer. Trained on 4,800 paired images from UIEB, EUVP, and LSUI, SFormer surpasses recent methods with a 3.1 dB gain in PSNR and 0.08 in SSIM, successfully restoring colors, textures, and contrast in underwater scenes.