HDR Reconstruction Boosting with Training-Free and Exposure-Consistent Diffusion

2026-02-23

AI Summary

This work addresses the challenge of reconstructing overexposed regions in single low dynamic range (LDR) to high dynamic range (HDR) image conversion, where information loss renders recovery particularly difficult. The authors propose a training-free diffusion-based enhancement method that leverages text-guided diffusion models combined with SDEdit refinement to synthesize plausible content for saturated areas. An iterative compensation mechanism is introduced to preserve luminance consistency across multiple exposures. Notably, this approach significantly improves both perceptual quality and quantitative metrics without modifying existing HDR reconstruction models. Extensive experiments on standard benchmarks and real-world scenes demonstrate its effectiveness in recovering natural details in overexposed regions while maintaining multi-exposure consistency and visual fidelity.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Single LDR to HDR reconstruction remains challenging for over-exposed regions where traditional methods often fail due to complete information loss. We present a training-free approach that enhances existing indirect and direct HDR reconstruction methods through diffusion-based inpainting. Our method combines text-guided diffusion models with SDEdit refinement to generate plausible content in over-exposed areas while maintaining consistency across multi-exposure LDR images. Unlike previous approaches requiring extensive training, our method seamlessly integrates with existing HDR reconstruction techniques through an iterative compensation mechanism that ensures luminance coherence across multiple exposures. We demonstrate significant improvements in both perceptual quality and quantitative metrics on standard HDR datasets and in-the-wild captures. Results show that our method effectively recovers natural details in challenging scenarios while preserving the advantages of existing HDR reconstruction pipelines. Project page: https://github.com/EusdenLin/HDR-Reconstruction-Boosting

Problem

• HDR reconstruction
• over-exposed regions
• information loss
• LDR to HDR
• dynamic range

Innovation

• training-free
• diffusion-based inpainting
• exposure-consistent
• HDR reconstruction
• SDEdit refinement