Multi-Exposure Image Fusion via Distilled 3D LUT Grid with Editable Mode

TL;DR

This paper introduces a real-time, high-resolution multi-exposure image fusion method using a distilled 3D LUT grid, enhanced by a teacher-student network to model uncertainty and an editable mode for adaptability.

Contribution

It presents a novel 3D LUT-based fusion approach with uncertainty modeling and an implicit representation for scenario-specific editing, suitable for ultra-high-definition images on resource-limited devices.

Findings

Achieves real-time UHD image fusion on resource-constrained devices.

Demonstrates high accuracy and efficiency compared to existing methods.

Provides an editable framework adaptable to various scenarios.

Abstract

With the rising imaging resolution of handheld devices, existing multi-exposure image fusion algorithms struggle to generate a high dynamic range image with ultra-high resolution in real-time. Apart from that, there is a trend to design a manageable and editable algorithm as the different needs of real application scenarios. To tackle these issues, we introduce 3D LUT technology, which can enhance images with ultra-high-definition (UHD) resolution in real time on resource-constrained devices. However, since the fusion of information from multiple images with different exposure rates is uncertain, and this uncertainty significantly trials the generalization power of the 3D LUT grid. To address this issue and ensure a robust learning space for the model, we propose using a teacher-student network to model the uncertainty on the 3D LUT grid.Furthermore, we provide an editable mode for the multi-exposure image fusion algorithm by using the implicit representation function to match the requirements in different scenarios. Extensive experiments demonstrate that our proposed method is highly competitive in efficiency and accuracy.