UHD-MFF: Shattering Barriers in Multi-Focus Ultra-High-Definition Image Fusion via Learnable Lookup Tables

2026-06-30

AI Summary

Existing methods for ultra-high-definition (UHD) multi-focus image fusion are hindered by data scarcity, limited model adaptability, and high deployment costs. To address these challenges, this work introduces the first large-scale UHD multi-focus fusion dataset and proposes the UMF-LUT framework, which enables multi-scale learnable lookup-table-based fusion through a coarse-to-fine strategy: a region-level decision module (C-LUT) guided by semantic and gradient cues, and a detail-level decision module (D-LUT) driven by Laplacian edge information. The proposed approach achieves extremely low computational overhead, supports real-time 4K inference, and surpasses state-of-the-art methods in both visual quality and quantitative metrics. Notably, it realizes real-time UHD multi-focus fusion on smartphones for the first time, significantly advancing practical deployment of this technology.

Abstract

With the advancement of imaging technology, ultra-high-definition images have become increasingly essential in modern visual applications. However, existing multi-focus image fusion remains largely confined to low-resolution images and faces three major barriers in UHD scenarios, namely data availability, model adaptability, and deployment feasibility, which severely hinder its practical application. To shatter these barriers, first, we propose the UHD-MFF dataset, the first large-scale ultra-high-resolution multi-focus fusion dataset. Second, we propose a scale-specialized lookup-table framework tailored for ultra-high-resolution images, termed as UMF-LUT. It consists of Coarse-Region Lookup Table (C-LUT) and Detail-Edge Lookup Table (D-LUT). Specifically, C-LUT performs joint queries of multiple gradient cues and semantic cues at low-resolution scales to enable region-level decision-making. Also, D-LUT operates at high-resolution scales, leveraging efficient Laplacian cues to provide complementary edge-level decision information. Such a design makes the model particularly well-suited for ultra-high-resolution multi-focus image fusion. Finally, it offers strong deployability with minimal computational overhead, enabling real-time 4K multi-focus fusion and showing promising potential for smartphone. Extensive experiments demonstrate that it outperforms SOTA methods in both visual fidelity and quantitative metrics. It effectively advances the development of multi-focus image fusion toward ultra-high-resolution imaging scenarios. The code is available at https://github.com/zyb5/UHD-MFF.

Problem

• multi-focus image fusion
• ultra-high-definition
• data availability
• model adaptability
• deployment feasibility

Innovation

• Learnable Lookup Tables
• Ultra-High-Definition Image Fusion
• Multi-Focus Fusion
• Scale-Specialized Framework
• Real-Time Deployment