# WaveFlow-UIE: A Single-Pipeline Wavelet-Domain Flow Model for Efficient Underwater Image Restoration

DevFeed: [WaveFlow-UIE: A Single-Pipeline Wavelet-Domain Flow Model for Efficient Underwater Image Restoration](<https://devfeed.tech/articles/waveflow-uie-a-single-pipeline-wavelet-domain-flow-model-for-efficient-underwater-image-restoration-132131.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15457>)

Author: Abdulaziz Alfaraj, Mohammed Al Naser, Hassan Al Nasser, Muzammil Behzad

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Image processing](<https://devfeed.tech/topics/image-processing.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Image](<https://devfeed.tech/topics/image.md>)

Tags: [ambient](<https://devfeed.tech/tags/ambient.md>), [ambient-light](<https://devfeed.tech/tags/ambient-light.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [color](<https://devfeed.tech/tags/color.md>), [components](<https://devfeed.tech/tags/components.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [domain](<https://devfeed.tech/tags/domain.md>), [domain-model](<https://devfeed.tech/tags/domain-model.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [flow](<https://devfeed.tech/tags/flow.md>), [flow-matching](<https://devfeed.tech/tags/flow-matching.md>), [image](<https://devfeed.tech/tags/image.md>), [inference](<https://devfeed.tech/tags/inference.md>), [light](<https://devfeed.tech/tags/light.md>), [lightweight](<https://devfeed.tech/tags/lightweight.md>), [model](<https://devfeed.tech/tags/model.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [restoration](<https://devfeed.tech/tags/restoration.md>)

## AI overview

WaveFlow-UIE restores underwater images by transforming degraded wavelet representations into clean ones in a single pipeline. It uses a discrete wavelet transform, lightweight estimates of transmission and ambient light, and a velocity network trained with flow matching; its losses also cover frequency-weighted reconstruction, perceptual quality, color consistency, and physics consistency. The abstract reports results across UIEB, LSUI, UFO-120, and EUVP, with competitive perceptual metrics and lower inference cost than a wavelet-based diffusion baseline while maintaining or improving quality on several paired benchmarks. On UIEB at 256x256, one flow step preserves nearly all the quality of five steps and adds about 4x speedup.

## Source excerpt

arXiv:2611.15457v1 Announce Type: cross Abstract: Underwater image restoration remains challenging because color distortion, scattering, and detail loss affect low- and high-frequency image components differently. Recent diffusion-based methods have shown strong restoration quality, but often rely on multi-stage pipelines and costly iterative refinement. We present WaveFlow-UIE, a single-pipeline wavelet-domain flow model that directly learns to transform degraded underwater wavelet representations into clean ones. Our method applies a discrete wavelet transform, estimates lightweight physics-inspired degradation cues including transmission and ambient light, and conditions a velocity network trained with flow matching in wavelet space. The training objective combines flow-matching, frequency-weighted reconstruction, perceptual, color-consistency, and physics-consistency losses. Across UIEB, LSUI, UFO-120, and EUVP, WaveFlow-UIE achieves a strong quality-efficiency tradeoff, with especially competitive results on perceptual metrics. Relative to a strong wavelet-based diffusion baseline, the proposed model substantially reduces inference cost while maintaining or improving restoration quality on several paired benchmarks. We show that reducing the inference process from five flow steps to a single step preserves nearly all of the restoration quality while providing an additional speedup of about $4\times$ on UIEB at $256\times256$ resolution. This single-step result suggests that a single-pass wavelet-domain model is an effective alternative to heavier diffusion-style refinement.