Enhancing Underwater Light Field Images via Global Geometry-aware Diffusion Process

2026-01-29

AI Summary

This work addresses color distortion and structural blurring in underwater four-dimensional light field imaging caused by scattering and absorption. To this end, the authors propose GeoDiff-LF, a geometry-aware diffusion model built upon SD-Turbo. By explicitly modeling the spatial-angular structure of light fields, GeoDiff-LF introduces a geometry-aware U-Net architecture, a tensor decomposition–guided geometric prior loss, and an attention-convolution adapter combined with an optimized noise prediction sampling strategy. This approach represents the first effort to integrate global geometric information into the light field diffusion process. Experimental results demonstrate that GeoDiff-LF significantly outperforms existing methods in both visual fidelity and quantitative metrics, effectively enhancing color reproduction and structural clarity in underwater light field images.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

This work studies the challenging problem of acquiring high-quality underwater images via 4-D light field (LF) imaging. To this end, we propose GeoDiff-LF, a novel diffusion-based framework built upon SD-Turbo to enhance underwater 4-D LF imaging by leveraging its spatial-angular structure. GeoDiff-LF consists of three key adaptations: (1) a modified U-Net architecture with convolutional and attention adapters to model geometric cues, (2) a geometry-guided loss function using tensor decomposition and progressive weighting to regularize global structure, and (3) an optimized sampling strategy with noise prediction to improve efficiency. By integrating diffusion priors and LF geometry, GeoDiff-LF effectively mitigates color distortion in underwater scenes. Extensive experiments demonstrate that our framework outperforms existing methods across both visual fidelity and quantitative performance, advancing the state-of-the-art in enhancing underwater imaging. The code will be publicly available at https://github.com/linlos1234/GeoDiff-LF.

Problem

• underwater imaging
• light field
• image enhancement
• color distortion
• 4-D LF imaging

Innovation

• light field imaging
• diffusion model
• underwater image enhancement
• geometry-aware
• tensor decomposition