Bayesian Neural Networks for One-to-Many Mapping in Image Enhancement

TL;DR

This paper introduces a Bayesian neural network-based framework for image enhancement that effectively models the one-to-many mapping problem, producing diverse and high-quality enhanced images efficiently.

Contribution

The paper presents a novel BNN-DNN framework that captures data uncertainty and enables fast, diverse image enhancement outputs for challenging scenarios.

Findings

Effective in low-light and underwater image enhancement benchmarks.

Produces diverse outputs capturing data uncertainty.

Outperforms existing methods in quality and speed.

Abstract

In image enhancement tasks, such as low-light and underwater image enhancement, a degraded image can correspond to multiple plausible target images due to dynamic photography conditions. This naturally results in a one-to-many mapping problem. To address this, we propose a Bayesian Enhancement Model (BEM) that incorporates Bayesian Neural Networks (BNNs) to capture data uncertainty and produce diverse outputs. To enable fast inference, we introduce a BNN-DNN framework: a BNN is first employed to model the one-to-many mapping in a low-dimensional space, followed by a Deterministic Neural Network (DNN) that refines fine-grained image details. Extensive experiments on multiple low-light and underwater image enhancement benchmarks demonstrate the effectiveness of our method.