Bracket Diffusion:
HDR Image Generation by Consistent LDR Denoising

Eurographics 2025

Mojtaba Bemana¹     Thomas Leimkühler¹     Karol Myszkowski¹    
Hans-Peter Seidel¹     Tobias Ritschel²

¹ MPI Informatik    ² University College London

Abstract

We demonstrate generating HDR images using the concerted action of multiple black-box, pre-trained LDR image diffusion models. Relying on a pre-trained LDR generative diffusion models is vital as, first, there is no sufficiently large HDR image dataset available to re-train them, and, second, even if it was, re-training such models is impossible for most compute budgets. Instead, we seek inspiration from the HDR image capture literature that traditionally fuses sets of LDR images, called "exposure brackets'', to produce a single HDR image. We operate multiple denoising processes to generate multiple LDR brackets that together form a valid HDR result. The key to making this work is to introduce a consistency term into the diffusion process to couple the brackets such that they agree across the exposure range they share while accounting for possible differences due to the quantization error. We demonstrate state-of-the-art unconditional and conditional or restoration-type (LDR2HDR) generative modeling results, yet in HDR.

Materials

• Paper
• Paper (arXiv)
• Supplementary PDF
• Source Code
• Results in HDR format (All the result figures in the paper)
• Evaluation set (Table 2) (input LDR, GT HDR, and our HDR results using 5 brackets)
• Evaluation set (Table 5) (input LDR, GT HDR, and our HDR results using 3, 5 ,and 7 brackets)
• Interactive comparison (Ours, MaskHDR, HDRCNN, GlowGan, and Reference)

Imprint / Data Protection