Dual-Output Multi-Exposure HDR Reconstruction via SDR Fusion and Gain Map Inverse Tone Mapping

2026-08-06

AI Summary

This work addresses the challenge of balancing perceptual quality and dynamic range consistency in multi-exposure HDR reconstruction by proposing the DOME-HDR framework. The method employs a LoRA-finetuned latent diffusion model to generate a perceptually optimized SDR image from three bracketed LDR inputs and introduces a dual-path cross-attention fusion module to effectively integrate structural and chromatic information from over- and under-exposed regions. This synthesized SDR image then guides a novel HDR Prior-Guided Gain Map (HPGM) network to predict spatially varying gain maps for high-quality HDR reconstruction. Extensive experiments demonstrate that the proposed approach outperforms state-of-the-art methods on the Kalantari, Tel, and Challenge123 datasets across both full-reference and no-reference metrics, while ablation studies confirm the effectiveness of each core component.

Abstract

We propose DOME-HDR, a dual-output multi-exposure HDR reconstruction framework that jointly produces a perceptually balanced SDR image and a consistent HDR image via gain map inverse tone mapping. Given three bracketed LDR inputs, DOME-HDR first synthesizes a base SDR using a LoRA-adapted latent diffusion model. A dual cross-attention fusion module injects complementary structural and color cues from the under- and over-exposed images while anchoring on the mid exposure for stability. The synthesized SDR then guides HPGM, our HDR Prior-guided Gain Map network, to predict a spatially varying gain map for reliable dynamic-range expansion. We evaluate on Kalantari, Tel, and Challenge123 using both full-reference and no-reference metrics, where DOME-HDR achieves state-of-the-art HDR reconstruction quality; ablations further confirm the effectiveness of dual cross-attention and SDR-guided gain map estimation.

Problem

• HDR reconstruction
• multi-exposure
• SDR fusion
• gain map
• inverse tone mapping

Innovation

• Dual-Output HDR
• Gain Map Inverse Tone Mapping
• LoRA-Adapted Diffusion Model
• Dual Cross-Attention Fusion
• SDR-Guided HDR Reconstruction