Physics Informed Capsule Enhanced Variational AutoEncoder for Underwater Image Enhancement

TL;DR

This paper introduces a physics-guided, capsule-enhanced variational autoencoder that significantly improves underwater image quality by combining physical models with deep learning, achieving state-of-the-art results efficiently.

Contribution

It proposes a dual-stream architecture integrating physical models with capsule clustering for underwater image enhancement, offering parameter-free, high-quality results with reduced computational cost.

Findings

+0.5dB PSNR over existing methods

Requires only one-third of the computational complexity

+1dB PSNR improvement at similar computational budgets

Abstract

We present a novel dual-stream architecture that achieves state-of-the-art underwater image enhancement by explicitly integrating the Jaffe-McGlamery physical model with capsule clustering-based feature representation learning. Our method simultaneously estimates transmission maps and spatially-varying background light through a dedicated physics estimator while extracting entity-level features via capsule clustering in a parallel stream. This physics-guided approach enables parameter-free enhancement that respects underwater formation constraints while preserving semantic structures and fine-grained details. Our approach also features a novel optimization objective ensuring both physical adherence and perceptual quality across multiple spatial frequencies. To validate our approach, we conducted extensive experiments across six challenging benchmarks. Results demonstrate consistent improvements of $+0.5$dB PSNR over the best existing methods while requiring only one-third of their computational complexity (FLOPs), or alternatively, more than $+1$dB PSNR improvement when compared to methods with similar computational budgets. Code and data \textit{will} be available at https://github.com/iN1k1/.