mlsanityData to 14 Oct 2026

Bracket Diffusion: HDR Image Generation by Consistent LDR Denoising

arXiv 2405.14304 ↗First posted 23 May 2024cs.GR cs.CV eess.IVcited by 15

On the map: Diffusion models → Inverse problems and image restoration

Abstract (arXiv)

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.

Key facts. Cited by 15 papers in the map (data through 14 Oct 2026). Placed in Inverse problems and image restoration, one of 14 directions in Diffusion models. Builds on 46 of its 78 references that are in the map. First posted on arXiv 23 May 2024.

Builds on

46 of 78 references.

Closest ancestors · 8

Same direction, newest first

  1. 2023Towards High-quality HDR Deghosting with Conditional Diffusion Models2 Nov
  2. 2023ExposureDiffusion: Learning to Expose for Low-light Image Enhancement15 Jul
  3. 2023Single Image LDR to HDR Conversion using Conditional Diffusion6 Jul
  4. 2023Solving Linear Inverse Problems Provably via Posterior Sampling with Latent Diffusion Models2 Jul
  5. 2023Denoising Diffusion Models for Plug-and-Play Image Restoration15 May
  6. 2023Generative Diffusion Prior for Unified Image Restoration and Enhancement3 Apr
  7. 2022Diffusion Posterior Sampling for General Noisy Inverse Problems29 Sep
  8. 2022Denoising Diffusion Restoration Models27 Jan

Other references · 38

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  5. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  6. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  7. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,614
  8. 2022DreamFusion: Text-to-3D using 2D Diffusioncited by 3,025
  9. 2021Image Super-Resolution via Iterative Refinementcited by 1,511
  10. 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,430
  11. 2022Diffusion Models: A Comprehensive Survey of Methods and Applicationscited by 1,114
  12. 2023MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generationcited by 536
26 more
  1. 2022StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasetscited by 492
  2. 2018Learning to See in the Darkcited by 460
  3. 2023DreamSim: Learning New Dimensions of Human Visual Similarity using Synthetic Datacited by 337
  4. 2021NeRF in the Dark: High Dynamic Range View Synthesis from Noisy Raw Imagescited by 336
  5. 2017HDR image reconstruction from a single exposure using deep CNNscited by 167
  6. 2020Single-Image HDR Reconstruction by Learning to Reverse the Camera Pipelinecited by 132
  7. 2019Attention-guided Network for Ghost-free High Dynamic Range Imagingcited by 117
  8. 2021HDR-NeRF: High Dynamic Range Neural Radiance Fieldscited by 114
  9. 2023SyncDiffusion: Coherent Montage via Synchronized Joint Diffusionscited by 110
  10. 2017Deep High Dynamic Range Imaging with Large Foreground Motionscited by 99
  11. 2018ExpandNet: A Deep Convolutional Neural Network for High Dynamic Range Expansion from Low Dynamic Range Contentcited by 89
  12. 2020HDR-GAN: HDR Image Reconstruction from Multi-Exposed LDR Images with Large Motionscited by 80
  13. 2022Any-resolution Training for High-resolution Image Synthesiscited by 72
  14. 2020Single Image HDR Reconstruction Using a CNN with Masked Features and Perceptual Losscited by 71
  15. 2023Mixture of Diffusers for scene composition and high resolution image generationcited by 56
  16. 2021Deep Learning for HDR Imaging: State-of-the-Art and Future Trendscited by 45
  17. 2023HDR-VDP-3: A multi-metric for predicting image differences, quality and contrast distortions in high dynamic range and regular contentcited by 41
  18. 2022HDR-Plenoxels: Self-Calibrating High Dynamic Range Radiance Fieldscited by 40
  19. 2018Deep Chain HDRI: Reconstructing a High Dynamic Range Image from a Single Low Dynamic Range Imagecited by 25
  20. 2022Text2Light: Zero-Shot Text-Driven HDR Panorama Generationcited by 24
  21. 2022GlowGAN: Unsupervised Learning of HDR Images from LDR Images in the Wildcited by 21
  22. 2020Low-light Image Restoration with Short- and Long-exposure Raw Pairscited by 18
  23. 2021Luminance Attentive Networks for HDR Image and Panorama Reconstructioncited by 18
  24. 2021UPHDR-GAN: Generative Adversarial Network for High Dynamic Range Imaging with Unpaired Datacited by 14
  25. 2023Learning Continuous Exposure Value Representations for Single-Image HDR Reconstructioncited by 11
  26. 2018LSD2 -- Joint Denoising and Deblurring of Short and Long Exposure Images with CNNscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 2; 2026: 12.

15 papers.

Sort:

By area:

  1. 2024LEDiff: Latent Exposure Diffusion for HDR Generationcited by 22
  2. 2025AIM 2025 challenge on Inverse Tone Mapping Report: Methods and Resultscited by 8
  3. 2026X2HDR: HDR Image Generation in a Perceptually Uniform Spacecited by 7
  4. 2026DiffHDR: Re-Exposing LDR Videos with Video Diffusion Modelscited by 3
  5. 2026HDR Video Generation via Latent Alignment with Logarithmic Encodingcited by 2
  6. 2026VidSplat: Gaussian Splatting Reconstruction with Geometry-Guided Video Diffusion Priorscited by 1
  7. 2025PhysHDR: When Lighting Meets Materials and Scene Geometry in HDR Reconstructioncited by 0
  8. 2026HDR Reconstruction Boosting with Training-Free and Exposure-Consistent Diffusioncited by 0
  9. 2026RawGen: Learning Camera Raw Image Generationcited by 0
  10. 2026Linear Image Generation by Synthesizing Exposure Bracketscited by 0
  11. 2026LatentHDR: Decoupling Exposure from Diffusion via Conditional Latent-to-Latent Mapping for Text/Image-to-Panoramic HDRcited by 0
  12. 2026Single-Shot HDR Recovery via a Video Diffusion Priorcited by 0
3 more
  1. 2026Coarse-to-fine Framework for Generative MEF via Implicit Neural Representationcited by 0
  2. 2026LumaGuide: Distribution Shaping for Training-Free HDR Generation in Diffusion Modelscited by 0
  3. 2026Recurrent Dynamic Range Extensioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Diffusion-Promoted HDR Video Reconstruction
  2. 2022ADIR: Adaptive Diffusion for Image Reconstruction
  3. 2026ExpoCM: Exposure-Aware One-Step Generative Single-Image HDR Reconstruction
  4. 2025GMODiff: One-Step Gain Map Refinement with Diffusion Priors for HDR Reconstruction
  5. 2023Faster Diffusion: Rethinking the Role of the Encoder for Diffusion Model Inference
  6. 2023Simple diffusion: End-to-end diffusion for high resolution images
  7. 2025LowDiff: Efficient Diffusion Sampling with Low-Resolution Condition
  8. 2024DiffuseHigh: Training-free Progressive High-Resolution Image Synthesis through Structure Guidance