Hybrid-Supervised Dual-Search: Leveraging Automatic Learning for Loss-free Multi-Exposure Image Fusion

TL;DR

This paper introduces HSDS-MEF, a novel dual-search framework that automatically designs network structures and loss functions for multi-exposure image fusion, significantly improving image quality and reducing artifacts.

Contribution

It proposes a bi-level optimization scheme with a dual research mechanism and hybrid supervised contrast constraint for automatic network and loss function design in MEF.

Findings

Achieves 10.61% improvement in VIF for general scenarios

Attains 4.38% improvement in VIF for no-reference scenarios

Produces images with high contrast, rich details, and accurate colors

Abstract

Multi-exposure image fusion (MEF) has emerged as a prominent solution to address the limitations of digital imaging in representing varied exposure levels. Despite its advancements, the field grapples with challenges, notably the reliance on manual designs for network structures and loss functions, and the constraints of utilizing simulated reference images as ground truths. Consequently, current methodologies often suffer from color distortions and exposure artifacts, further complicating the quest for authentic image representation. In addressing these challenges, this paper presents a Hybrid-Supervised Dual-Search approach for MEF, dubbed HSDS-MEF, which introduces a bi-level optimization search scheme for automatic design of both network structures and loss functions. More specifically, we harnesses a unique dual research mechanism rooted in a novel weighted structure refinement architecture search. Besides, a hybrid supervised contrast constraint seamlessly guides and integrates with searching process, facilitating a more adaptive and comprehensive search for optimal loss functions. We realize the state-of-the-art performance in comparison to various competitive schemes, yielding a 10.61% and 4.38% improvement in Visual Information Fidelity (VIF) for general and no-reference scenarios, respectively, while providing results with high contrast, rich details and colors.