mlsanityData to 14 Oct 2026

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

arXiv 2508.18664 ↗First posted 26 Aug 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 36 of its 84 references that are in the map. First posted on arXiv 26 Aug 2025.

Builds on

36 of 84 references.

References · 36

  1. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  2. 2019An Underwater Image Enhancement Benchmark Dataset and Beyondcited by 253
  3. 2019Fast Underwater Image Enhancement for Improved Visual Perceptioncited by 183
  4. 2022Ultra-High-Definition Low-Light Image Enhancement: A Benchmark and Transformer-Based Methodcited by 174
  5. 2021Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embeddingcited by 134
  6. 2023Explicit Visual Prompting for Low-Level Structure Segmentationscited by 114
  7. 2021U-shape Transformer for Underwater Image Enhancementcited by 112
  8. 2018Enhancing Underwater Imagery using Generative Adversarial Networkscited by 95
  9. 2023Contrastive Semi-supervised Learning for Underwater Image Restoration via Reliable Bankcited by 71
  10. 2024Seeing Clearly by Layer Two: Enhancing Attention Heads to Alleviate Hallucination in LVLMscited by 52
  11. 2023PUGAN: Physical Model-Guided Underwater Image Enhancement Using GAN with Dual-Discriminatorscited by 49
  12. 2021Domain Adaptation for Underwater Image Enhancementcited by 35
24 more
  1. 2023CoordFill: Efficient High-Resolution Image Inpainting via Parameterized Coordinate Queryingcited by 34
  2. 2023Toward Sufficient Spatial-Frequency Interaction for Gradient-aware Underwater Image Enhancementcited by 27
  3. 2024Depth-aware Test-Time Training for Zero-shot Video Object Segmentationcited by 27
  4. 2024DOPRA: Decoding Over-accumulation Penalization and Re-allocation in Specific Weighting Layercited by 27
  5. 2023SyreaNet: A Physically Guided Underwater Image Enhancement Framework Integrating Synthetic and Real Imagescited by 23
  6. 2020SiENet: Siamese Expansion Network for Image Extrapolationcited by 18
  7. 2025Underwater Image Restoration Through a Prior Guided Hybrid Sense Approach and Extensive Benchmark Analysiscited by 17
  8. 2024Image Fusion for Cross-Domain Sequential Recommendationcited by 17
  9. 2022A Wavelet-based Dual-stream Network for Underwater Image Enhancementcited by 16
  10. 2023A Generalized Physical-knowledge-guided Dynamic Model for Underwater Image Enhancementcited by 16
  11. 2024Prompt Categories Cluster for Weakly Supervised Semantic Segmentationcited by 16
  12. 2024Image Augmentation Agent for Weakly Supervised Semantic Segmentationcited by 15
  13. 2024SMAFormer: Synergistic Multi-Attention Transformer for Medical Image Segmentationcited by 14
  14. 2025MCA-LLaVA: Manhattan Causal Attention for Reducing Hallucination in Large Vision-Language Modelscited by 12
  15. 2024Test-Time Intensity Consistency Adaptation for Shadow Detectioncited by 11
  16. 2023Top-K Pooling with Patch Contrastive Learning for Weakly-Supervised Semantic Segmentationcited by 10
  17. 2023UWFormer: Underwater Image Enhancement via a Semi-Supervised Multi-Scale Transformercited by 9
  18. 2024ForgeryTTT: Zero-Shot Image Manipulation Localization with Test-Time Trainingcited by 8
  19. 2022Semantic-aware Texture-Structure Feature Collaboration for Underwater Image Enhancementcited by 7
  20. 2025AdaToken-3D: Dynamic Spatial Gating for Efficient 3D Large Multimodal-Models Reasoningcited by 5
  21. 2025CMAMRNet: A Contextual Mask-Aware Network Enhancing Mural Restoration Through Comprehensive Mask Guidancecited by 3
  22. 2024UIE-UnFold: Deep Unfolding Network with Color Priors and Vision Transformer for Underwater Image Enhancementcited by 2
  23. 2024GRE^2-MDCL: Graph Representation Embedding Enhanced via Multidimensional Contrastive Learningcited by 2
  24. 2024Lagrange Duality and Compound Multi-Attention Transformer for Semi-Supervised Medical Image Segmentationcited by 2

Built on by

Not cited yet.