mlsanityData to 14 Oct 2026

Dancing under the stars: video denoising in starlight

arXiv 2204.04210 ↗First posted 8 Apr 2022cs.CV eess.IVcited by 47

Not on the map. Why?

Abstract (arXiv)

Imaging in low light is extremely challenging due to low photon counts. Using sensitive CMOS cameras, it is currently possible to take videos at night under moonlight (0.05-0.3 lux illumination). In this paper, we demonstrate photorealistic video under starlight (no moon present, <0.001 lux) for the first time. To enable this, we develop a GAN-tuned physics-based noise model to more accurately represent camera noise at the lowest light levels. Using this noise model, we train a video denoiser using a combination of simulated noisy video clips and real noisy still images. We capture a 5-10 fps video dataset with significant motion at approximately 0.6-0.7 millilux with no active illumination. Comparing against alternative methods, we achieve improved video quality at the lowest light levels, demonstrating photorealistic video denoising in starlight for the first time.

Key facts. Cited by 47 papers in the map (data through 14 Oct 2026). Builds on 30 of its 60 references that are in the map. First posted on arXiv 8 Apr 2022.

Builds on

30 of 60 references.

References · 30

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  4. 2022Generative Adversarial Networkscited by 5,627
  5. 2017Improved Training of Wasserstein GANscited by 2,235
  6. 2016Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoisingcited by 1,358
  7. 2017Deep Image Priorcited by 1,098
  8. 2018Noise2Noise: Learning Image Restoration without Clean Datacited by 533
  9. 2018Learning to See in the Darkcited by 460
  10. 2018Noise2Void - Learning Denoising from Single Noisy Imagescited by 364
  11. 2019Noise2Self: Blind Denoising by Self-Supervisioncited by 276
  12. 2018Unprocessing Images for Learned Raw Denoisingcited by 250
18 more
  1. 2017Benchmarking Denoising Algorithms with Real Photographscited by 209
  2. 2017Burst Denoising with Kernel Prediction Networkscited by 157
  3. 2020A Physics-based Noise Formation Model for Extreme Low-light Raw Denoisingcited by 138
  4. 2015MOTChallenge 2015: Towards a Benchmark for Multi-Target Trackingcited by 133
  5. 2019FastDVDnet: Towards Real-Time Deep Video Denoising Without Flow Estimationcited by 128
  6. 2019Noise Flow: Noise Modeling with Conditional Normalizing Flowscited by 108
  7. 2021Rethinking Noise Synthesis and Modeling in Raw Denoisingcited by 62
  8. 2019Handheld Mobile Photography in Very Low Lightcited by 56
  9. 2019Probabilistic Noise2Void: Unsupervised Content-Aware Denoisingcited by 51
  10. 2017Deep Burst Denoisingcited by 49
  11. 2019ViDeNN: Deep Blind Video Denoisingcited by 40
  12. 2021Patch Craft: Video Denoising by Deep Modeling and Patch Matchingcited by 38
  13. 2014RENOIR - A Dataset for Real Low-Light Image Noise Reductioncited by 36
  14. 2018Model-blind Video Denoising Via Frame-to-frame Trainingcited by 34
  15. 2020Learning Camera-Aware Noise Modelscited by 32
  16. 2014High-level numerical simulations of noise in CCD and CMOS photosensors: review and tutorialcited by 23
  17. 2019Fully Unsupervised Probabilistic Noise2Voidcited by 14
  18. 2018Class-Aware Fully-Convolutional Gaussian and Poisson Denoisingcited by 12

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 6; 2023: 12; 2024: 9; 2025: 14; 2026: 6.

47 papers.

Sort:

By area:

  1. 2023Physics-Informed Computer Vision: A Review and Perspectivescited by 28
  2. 2022DynamicISP: Dynamically Controlled Image Signal Processor for Image Recognitioncited by 23
  3. 2023Diffusion in the Dark: A Diffusion Model for Low-Light Text Recognitioncited by 20
  4. 2023Towards General Low-Light Raw Noise Synthesis and Modelingcited by 19
  5. 2024RAW-Adapter: Adapting Pre-trained Visual Model to Camera RAW Imagescited by 19
  6. 2022Rawgment: Noise-Accounted RAW Augmentation Enables Recognition in a Wide Variety of Environmentscited by 18
  7. 2022Low-Light Image and Video Enhancement: A Comprehensive Survey and Beyondcited by 17
  8. 2025AIM 2025 Low-light RAW Video Denoising Challenge: Dataset, Methods and Resultscited by 14
  9. 2024DualDn: Dual-domain Denoising via Differentiable ISPcited by 11
  10. 2023Learning Physics-Informed Noise Models from Dark Frames for Low-Light Raw Image Denoisingcited by 10
  11. 2022Shakes on a Plane: Unsupervised Depth Estimation from Unstabilized Photographycited by 8
  12. 2023SVNR: Spatially-variant Noise Removal with Denoising Diffusioncited by 8
35 more
  1. 2024BVI-RLV: A Fully Registered Dataset for Low-Light Video Enhancementcited by 8
  2. 2026Denoising the Deep Sky: Physics-Based CCD Noise Formation for Astronomical Imagingcited by 8
  3. 2022Modular Degradation Simulation and Restoration for Under-Display Cameracited by 7
  4. 2023Dual-Camera Joint Deblurring-Denoisingcited by 6
  5. 2023Recurrent Spike-based Image Restoration under General Illuminationcited by 5
  6. 2023DiffuseRAW: End-to-End Generative RAW Image Processing for Low-Light Imagescited by 4
  7. 2024SeNM-VAE: Semi-Supervised Noise Modeling with Hierarchical Variational Autoencodercited by 4
  8. 2025Dark Noise Diffusion: Noise Synthesis for Low-Light Image Denoisingcited by 4
  9. 2023A Generative Model for Digital Camera Noise Synthesiscited by 3
  10. 2023Learned, uncertainty-driven adaptive acquisition for photon-efficient scanning microscopycited by 3
  11. 2024Event Cameras Meet SPADs for High-Speed, Low-Bandwidth Imagingcited by 3
  12. 2024Quanta Video Restorationcited by 3
  13. 2024Zero-Shot Low Light Image Enhancement with Diffusion Priorcited by 3
  14. 2025Towards a General-Purpose Zero-Shot Synthetic Low-Light Image and Video Pipelinecited by 3
  15. 2023Contrastive Learning for Low-light Raw Denoisingcited by 2
  16. 2023Make Explicit Calibration Implicit: Calibrate Denoiser Instead of the Noise Modelcited by 2
  17. 2025Digitally Prototype Your Eye Tracker: Simulating Hardware Performance using 3D Synthetic Datacited by 2
  18. 2025RAW-Adapter: Adapting Pre-trained Visual Model to Camera RAW Images and A Benchmarkcited by 2
  19. 2025AIM 2025 Challenge on Real-World RAW Image Denoisingcited by 2
  20. 20252-Shots in the Dark: Low-Light Denoising with Minimal Data Acquisitioncited by 2
  21. 2022Learning to Kindle the Starlightcited by 1
  22. 2024Real-Time Convolutional Neural Network-Based Star Detection and Centroiding Method for CubeSat Star Trackercited by 1
  23. 2025Unsupervised Methods for Video Quality Improvement: A Survey of Restoration and Enhancement Techniquescited by 1
  24. 2025ELVIS: Enhance Low-Light for Video Instance Segmentation in the Darkcited by 1
  25. 2026UniISP: A Unified ISP Framework for Both Human and Machine Visioncited by 1
  26. 2026MoCRA: Mixture of Compositional Rank-1 Atoms for 4K All-in-One Video Restorationcited by 1
  27. 2024Adaptive Domain Learning for Cross-domain Image Denoisingcited by 0
  28. 2025Spectrum from Defocus: Fast Spectral Imaging with Chromatic Focal Stackcited by 0
  29. 2025CIM-NET: A Video Denoising Deep Neural Network Model Optimized for Computing-in-Memory Architecturescited by 0
  30. 2025Image Restoration Learning via Noisy Supervision in the Fourier Domaincited by 0
  31. 2025Physics-Guided Rectified Flow for Low-light RAW Image Enhancementcited by 0
  32. 2025Dynamic Weight-based Temporal Aggregation for Low-light Video Enhancement Under Extreme Noisecited by 0
  33. 2026UHD-GPGNet: UHD Video Denoising via Gaussian-Process-Guided Local Spatio-Temporal Modelingcited by 0
  34. 2026Poisson2Gaussian: Noise Gaussianization to Enhance Image Denoisingcited by 0
  35. 2026Practical Noise Modeling for SPAD Intensity Imagingcited by 0