# Retinex Meets Language: A Physics-Semantics-Guided Underwater Image Enhancement Network

arXiv: https://arxiv.org/abs/2603.07076 · HTML: https://mlsanity.com/paper/2603.07076/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 48 references that are in the map.
- First posted on arXiv 7 Mar 2026.

## Builds on (most cited first, 21 of 21)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [VQA: Visual Question Answering](https://mlsanity.com/paper/1505.00468/) (2015)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [An Underwater Image Enhancement Benchmark Dataset and Beyond](https://mlsanity.com/paper/1901.05495/) (2019)
- [Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embedding](https://mlsanity.com/paper/2104.13015/) (2021)
- [U-shape Transformer for Underwater Image Enhancement](https://mlsanity.com/paper/2111.11843/) (2021)
- [Iterative Prompt Learning for Unsupervised Backlit Image Enhancement](https://mlsanity.com/paper/2303.17569/) (2023)
- [Underwater Single Image Color Restoration Using Haze-Lines and a New Quantitative Dataset](https://mlsanity.com/paper/1811.01343/) (2018)
- [Underwater Ranker: Learn Which Is Better and How to Be Better](https://mlsanity.com/paper/2208.06857/) (2022)
- [Retinexmamba: Retinex-based Mamba for Low-light Image Enhancement](https://mlsanity.com/paper/2405.03349/) (2024)
- [Uncertainty Inspired Underwater Image Enhancement](https://mlsanity.com/paper/2207.09689/) (2022)
- [SGUIE-Net: Semantic Attention Guided Underwater Image Enhancement with Multi-Scale Perception](https://mlsanity.com/paper/2201.02832/) (2022)
- [RAVE: Residual Vector Embedding for CLIP-Guided Backlit Image Enhancement](https://mlsanity.com/paper/2404.01889/) (2024)
- [UIF: An Objective Quality Assessment for Underwater Image Enhancement](https://mlsanity.com/paper/2205.09392/) (2022)
- [UWFormer: Underwater Image Enhancement via a Semi-Supervised Multi-Scale Transformer](https://mlsanity.com/paper/2310.20210/) (2023)
- [Underwater Image Enhancement by Diffusion Model with Customized CLIP-Classifier](https://mlsanity.com/paper/2405.16214/) (2024)
- [HazeCLIP: Towards Language Guided Real-World Image Dehazing](https://mlsanity.com/paper/2407.13719/) (2024)
- [A Semi-supervised Physics-Aware Triple-Stream Underwater Image Enhancement Network](https://mlsanity.com/paper/2307.11470/) (2023)

## Abstract (arXiv)

Underwater images often suffer from severe degradation caused by light absorption and scattering, leading to color distortion, low contrast and reduced visibility. Existing Underwater Image Enhancement (UIE) methods can be divided into two categories, i.e., prior-based and learning-based methods. The former rely on rigid physical assumptions that limit the adaptability, while the latter often face data scarcity and weak generalization. To address these issues, we propose a Physics-Semantics-Guided Underwater Image Enhancement Network (PSG-UIENet), which couples the Retinex-grounded illumination correction with the language-informed guidance. This network comprises a Prior-Free Illumination Estimator and a Semantics-Guided Image Restorer. In particular, the restorer leverages the textual descriptions generated by the Contrastive Language-Image Pre-training (CLIP) model to inject high-level semantics for perceptually meaningful guidance. Since multimodal UIE data sets are not publicly available, we also construct a large-scale image-text UIE data set, namely, LUIQD-TD, which contains 6,418 image-reference-text triplets. To explicitly measure and optimize semantic consistency between textual descriptions and images, we further design an Image-Text Semantic Similarity (ITSS) loss function. To our knowledge, this study makes the first effort to introduce both textual guidance and the multimodal data set into UIE tasks. Extensive experiments on our data set and four publicly available data sets demonstrate that the proposed PSG-UIENet achieves superior or comparable performance against fifteen state-of-the-art methods.