mlsanityData to 7 Oct 2026

Rethinking Noise Synthesis and Modeling in Raw Denoising

arXiv 2110.04756 ↗First posted 10 Oct 2021eess.IV cs.CVcited by 62

Not on the map. Why?

Abstract (arXiv)

The lack of large-scale real raw image denoising dataset gives rise to challenges on synthesizing realistic raw image noise for training denoising models. However, the real raw image noise is contributed by many noise sources and varies greatly among different sensors. Existing methods are unable to model all noise sources accurately, and building a noise model for each sensor is also laborious. In this paper, we introduce a new perspective to synthesize noise by directly sampling from the sensor's real noise. It inherently generates accurate raw image noise for different camera sensors. Two efficient and generic techniques: pattern-aligned patch sampling and high-bit reconstruction help accurate synthesis of spatial-correlated noise and high-bit noise respectively. We conduct systematic experiments on SIDD and ELD datasets. The results show that (1) our method outperforms existing methods and demonstrates wide generalization on different sensors and lighting conditions. (2) Recent conclusions derived from DNN-based noise modeling methods are actually based on inaccurate noise parameters. The DNN-based methods still cannot outperform physics-based statistical methods.

Key facts. Cited by 62 papers in the map (data through 7 Oct 2026). Builds on 11 of its 32 references that are in the map. First posted on arXiv 10 Oct 2021.

Builds on

11 of 32 references.

References · 11

  1. 2016Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoisingcited by 1,352
  2. 2018Noise2Noise: Learning Image Restoration without Clean Datacited by 526
  3. 2018Learning to See in the Darkcited by 459
  4. 2018Unprocessing Images for Learned Raw Denoisingcited by 250
  5. 2017Burst Denoising with Kernel Prediction Networkscited by 156
  6. 2020A Physics-based Noise Formation Model for Extreme Low-light Raw Denoisingcited by 138
  7. 2019Noise Flow: Noise Modeling with Conditional Normalizing Flowscited by 108
  8. 2020Dual Adversarial Network: Toward Real-world Noise Removal and Noise Generationcited by 101
  9. 2019GRDN:Grouped Residual Dense Network for Real Image Denoising and GAN-based Real-world Noise Modelingcited by 33
  10. 2020Learning Camera-Aware Noise Modelscited by 32
  11. 2014High-level numerical simulations of noise in CCD and CMOS photosensors: review and tutorialcited by 23

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 10; 2023: 17; 2024: 9; 2025: 14; 2026: 11.

62 papers.

Sort:

By area:

  1. 2023Image Denoising: The Deep Learning Revolution and Beyond -- A Survey Paper --cited by 71
  2. 2022Realistic Blur Synthesis for Learning Image Deblurringcited by 51
  3. 2022Dancing under the stars: video denoising in starlightcited by 47
  4. 2021IDR: Self-Supervised Image Denoising via Iterative Data Refinementcited by 36
  5. 2023KBNet: Kernel Basis Network for Image Restorationcited by 33
  6. 2023Spatially Adaptive Self-Supervised Learning for Real-World Image Denoisingcited by 33
  7. 2023ExposureDiffusion: Learning to Expose for Low-light Image Enhancementcited by 29
  8. 2022Modeling sRGB Camera Noise with Normalizing Flowscited by 28
  9. 2022Learnability Enhancement for Low-light Raw Denoising: Where Paired Real Data Meets Noise Modelingcited by 23
  10. 2023A Unified Conditional Framework for Diffusion-based Image Restorationcited by 22
  11. 2023Towards General Low-Light Raw Noise Synthesis and Modelingcited by 19
  12. 2023Raw Image Reconstruction with Learned Compact Metadatacited by 18
38 more
  1. 2022Noise2NoiseFlow: Realistic Camera Noise Modeling without Clean Imagescited by 17
  2. 2022Efficient Burst Raw Denoising with Variance Stabilization and Multi-frequency Denoising Networkcited by 15
  3. 2022Estimating Fine-Grained Noise Model via Contrastive Learningcited by 14
  4. 2023BSRAW: Improving Blind RAW Image Super-Resolutioncited by 11
  5. 2025Learning to See in the Extremely Darkcited by 11
  6. 2023Learning Physics-Informed Noise Models from Dark Frames for Low-Light Raw Image Denoisingcited by 10
  7. 2025Time-Aware Auto White Balance in Mobile Photographycited by 10
  8. 2023Beyond Learned Metadata-based Raw Image Reconstructioncited by 9
  9. 2024CMTA: Cross-Modal Temporal Alignment for Event-guided Video Deblurringcited by 9
  10. 2023RViDeformer: Efficient Raw Video Denoising Transformer with a Larger Benchmark Datasetcited by 8
  11. 2026Denoising the Deep Sky: Physics-Based CCD Noise Formation for Astronomical Imagingcited by 8
  12. 2022Modular Degradation Simulation and Restoration for Under-Display Cameracited by 7
  13. 2023Visibility Constrained Wide-band Illumination Spectrum Design for Seeing-in-the-Darkcited by 7
  14. 2024Toward Efficient Deep Blind RAW Image Restorationcited by 7
  15. 2024Retinex-RAWMamba: Bridging Demosaicing and Denoising for Low-Light RAW Image Enhancementcited by 6
  16. 2024Temporal As a Plugin: Unsupervised Video Denoising with Pre-Trained Image Denoiserscited by 6
  17. 2023Recurrent Spike-based Image Restoration under General Illuminationcited by 5
  18. 2025Dark-ISP: Enhancing RAW Image Processing for Low-Light Object Detectioncited by 5
  19. 2023DiffuseRAW: End-to-End Generative RAW Image Processing for Low-Light Imagescited by 4
  20. 2024Self-Supervised Monocular Depth Estimation in the Dark: Towards Data Distribution Compensationcited by 4
  21. 2025Dark Noise Diffusion: Noise Synthesis for Low-Light Image Denoisingcited by 4
  22. 2025Beyond RGB: Adaptive Parallel Processing for RAW Object Detectioncited by 4
  23. 2025YOND: Practical Blind Raw Image Denoising Free from Camera-Specific Data Dependencycited by 4
  24. 2026RAW-Flow: Advancing RGB-to-RAW Image Reconstruction with Deterministic Latent Flow Matchingcited by 4
  25. 2023A Generative Model for Digital Camera Noise Synthesiscited by 3
  26. 2023Efficient HDR Reconstruction from Real-World Raw Imagescited by 3
  27. 2023Make Explicit Calibration Implicit: Calibrate Denoiser Instead of the Noise Modelcited by 2
  28. 2024Modeling Dual-Exposure Quad-Bayer Patterns for Joint Denoising and Deblurringcited by 2
  29. 2025TS-Diff: Two-Stage Diffusion Model for Low-Light RAW Image Enhancementcited by 2
  30. 2025AIM 2025 Challenge on Real-World RAW Image Denoisingcited by 2
  31. 20252-Shots in the Dark: Low-Light Denoising with Minimal Data Acquisitioncited by 2
  32. 2025Towards a Universal Image Degradation Model via Content-Degradation Disentanglementcited by 1
  33. 2026Why Low-Light Cameras Go Color Blind: Removing Color Bias in Raw Denoisingcited by 1
  34. 2022Towards Real-World Video Denosing: A Practical Video Denosing Dataset and Networkcited by 0
  35. 2022Learn to See Faster: Pushing the Limits of High-Speed Camera with Deep Underexposed Image Denoisingcited by 0
  36. 2024Real-time Noise Source Estimation of a Camera System from an Image and Metadatacited by 0
  37. 2024Adaptive Domain Learning for Cross-domain Image Denoisingcited by 0
  38. 2024Video Denoising in Fluorescence Guided Surgerycited by 0

50 most cited of 62 All 62 (JSON)