mlsanityData to 7 Oct 2026

ShadowDiffusion: When Degradation Prior Meets Diffusion Model for Shadow Removal

arXiv 2212.04711 ↗First posted 9 Dec 2022cs.CVcited by 102

On the map: Diffusion models → 3D, point cloud and scene generation

Abstract (arXiv)

Recent deep learning methods have achieved promising results in image shadow removal. However, their restored images still suffer from unsatisfactory boundary artifacts, due to the lack of degradation prior embedding and the deficiency in modeling capacity. Our work addresses these issues by proposing a unified diffusion framework that integrates both the image and degradation priors for highly effective shadow removal. In detail, we first propose a shadow degradation model, which inspires us to build a novel unrolling diffusion model, dubbed ShandowDiffusion. It remarkably improves the model's capacity in shadow removal via progressively refining the desired output with both degradation prior and diffusive generative prior, which by nature can serve as a new strong baseline for image restoration. Furthermore, ShadowDiffusion progressively refines the estimated shadow mask as an auxiliary task of the diffusion generator, which leads to more accurate and robust shadow-free image generation. We conduct extensive experiments on three popular public datasets, including ISTD, ISTD+, and SRD, to validate our method's effectiveness. Compared to the state-of-the-art methods, our model achieves a significant improvement in terms of PSNR, increasing from 31.69dB to 34.73dB over SRD dataset.

Key facts. Cited by 102 papers in the map (data through 7 Oct 2026). Placed in 3D, point cloud and scene generation, one of 14 directions in Diffusion models. Builds on 46 of its 84 references that are in the map. First posted on arXiv 9 Dec 2022.

Builds on

46 of 84 references.

Closest ancestors · 1

Same direction, newest first

  1. 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Models24 Jan

Other references · 45

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  4. 2020Denoising Diffusion Implicit Modelscited by 9,655
  5. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  6. 2022Generative Adversarial Networkscited by 5,603
  7. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  8. 2014Conditional Generative Adversarial Netscited by 2,152
  9. 2021Image Super-Resolution via Iterative Refinementcited by 1,506
  10. 2021Palette: Image-to-Image Diffusion Modelscited by 1,296
  11. 2022Denoising Diffusion Restoration Modelscited by 951
  12. 2021Uformer: A General U-Shaped Transformer for Image Restorationcited by 864
33 more
  1. 2018Deep Retinex Decomposition for Low-Light Enhancementcited by 720
  2. 2020Zero-Reference Deep Curve Estimation for Low-Light Image Enhancementcited by 541
  3. 2019EnlightenGAN: Deep Light Enhancement without Paired Supervisioncited by 466
  4. 2019Kindling the Darkness: A Practical Low-light Image Enhancercited by 355
  5. 2016Deep Joint Rain Detection and Removal from a Single Imagecited by 351
  6. 2020Learning Enriched Features for Real Image Restoration and Enhancementcited by 311
  7. 2020Retinex-inspired Unrolling with Cooperative Prior Architecture Search for Low-light Image Enhancementcited by 303
  8. 2020Deep Unfolding Network for Image Super-Resolutioncited by 253
  9. 2016Plug-and-Play ADMM for Image Restoration: Fixed Point Convergence and Applicationscited by 244
  10. 2017A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstructioncited by 231
  11. 2022Restoring Vision in Adverse Weather Conditions with Patch-Based Denoising Diffusion Modelscited by 215
  12. 2020SRFlow: Learning the Super-Resolution Space with Normalizing Flowcited by 209
  13. 2021Low-Light Image Enhancement with Normalizing Flowcited by 184
  14. 2017HDR image reconstruction from a single exposure using deep CNNscited by 167
  15. 2017Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removalcited by 160
  16. 2017DSLR-Quality Photos on Mobile Devices with Deep Convolutional Networkscited by 156
  17. 2019Towards Ghost-free Shadow Removal via Dual Hierarchical Aggregation Network and Shadow Matting GANcited by 102
  18. 2019Shadow Removal via Shadow Image Decompositioncited by 96
  19. 2019Mask-ShadowGAN: Learning to Remove Shadows from Unpaired Datacited by 91
  20. 2021Auto-Exposure Fusion for Single-Image Shadow Removalcited by 78
  21. 2022Unsupervised Night Image Enhancement: When Layer Decomposition Meets Light-Effects Suppressioncited by 73
  22. 2022DC-ShadowNet: Single-Image Hard and Soft Shadow Removal Using Unsupervised Domain-Classifier Guided Networkcited by 68
  23. 2019Attention Guided Low-light Image Enhancement with a Large Scale Low-light Simulation Datasetcited by 65
  24. 2017Direction-aware Spatial Context Features for Shadow Detectioncited by 56
  25. 2021From Shadow Generation to Shadow Removalcited by 54
  26. 2019ARGAN: Attentive Recurrent Generative Adversarial Network for Shadow Detection and Removalcited by 51
  27. 2022Towards Robust Rain Removal Against Adversarial Attacks: A Comprehensive Benchmark Analysis and Beyondcited by 48
  28. 2021CANet: A Context-Aware Network for Shadow Removalcited by 47
  29. 2018Image Correction via Deep Reciprocating HDR Transformationcited by 30
  30. 2021ReLLIE: Deep Reinforcement Learning for Customized Low-Light Image Enhancementcited by 26
  31. 2022Estimating Reflectance Layer from A Single Image: Integrating Reflectance Guidance and Shadow/Specular Aware Learningcited by 18
  32. 2022Enhancing Low-Light Images in Real World via Cross-Image Disentanglementcited by 4
  33. 2021FINO: Flow-based Joint Image and Noise Modelcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 19; 2024: 37; 2025: 26; 2026: 19.

102 papers.

Sort:

By area:

  1. 2024Diffusion Model-Based Image Editing: A Surveycited by 212
  2. 2023Low-Light Image Enhancement with Wavelet-based Diffusion Modelscited by 126
  3. 2023Diffusion Models for Image Restoration and Enhancement: A Comprehensive Surveycited by 108
  4. 2023AutoDIR: Automatic All-in-One Image Restoration with Latent Diffusioncited by 98
  5. 2024ObjectDrop: Bootstrapping Counterfactuals for Photorealistic Object Removal and Insertioncited by 71
  6. 2023Residual Denoising Diffusion Modelscited by 68
  7. 2024I2EBench: A Comprehensive Benchmark for Instruction-based Image Editingcited by 61
  8. 2024DMPlug: A Plug-in Method for Solving Inverse Problems with Diffusion Modelscited by 60
  9. 2024LightIt: Illumination Modeling and Control for Diffusion Modelscited by 56
  10. 2024Diff-Plugin: Revitalizing Details for Diffusion-based Low-level Taskscited by 47
  11. 2023Backdoor Attacks Against Deep Image Compression via Adaptive Frequency Triggercited by 40
  12. 2025Vision-Language Gradient Descent-driven All-in-One Deep Unfolding Networkscited by 37
38 more
  1. 2023ShadowFormer: Global Context Helps Image Shadow Removalcited by 35
  2. 2024Light the Night: A Multi-Condition Diffusion Framework for Unpaired Low-Light Enhancement in Autonomous Drivingcited by 35
  3. 2025ObjectMover: Generative Object Movement with Video Priorcited by 30
  4. 2023ExposureDiffusion: Learning to Expose for Low-light Image Enhancementcited by 29
  5. 2023Latent Feature-Guided Diffusion Models for Shadow Removalcited by 28
  6. 2024Teaching Tailored to Talent: Adverse Weather Restoration via Prompt Pool and Depth-Anything Constraintcited by 24
  7. 2025RORem: Training a Robust Object Remover with Human-in-the-Loopcited by 23
  8. 2023UrbanIR: Large-Scale Urban Scene Inverse Rendering from a Single Videocited by 22
  9. 2024AsyncDiff: Parallelizing Diffusion Models by Asynchronous Denoisingcited by 22
  10. 2023Towards Real-World Blind Face Restoration with Generative Diffusion Priorcited by 21
  11. 2023Resfusion: Denoising Diffusion Probabilistic Models for Image Restoration Based on Prior Residual Noisecited by 19
  12. 2024OmniSR: Shadow Removal under Direct and Indirect Lightingcited by 19
  13. 2022QuantArt: Quantizing Image Style Transfer Towards High Visual Fidelitycited by 18
  14. 2023Raw Image Reconstruction with Learned Compact Metadatacited by 18
  15. 2024Shadow Generation for Composite Image Using Diffusion modelcited by 17
  16. 2023LLDiffusion: Learning Degradation Representations in Diffusion Models for Low-Light Image Enhancementcited by 15
  17. 2023Improving the Stability and Efficiency of Diffusion Models for Content Consistent Super-Resolutioncited by 15
  18. 2024DP-IQA: Utilizing Diffusion Prior for Blind Image Quality Assessment in the Wildcited by 14
  19. 2024DiffRetouch: Using Diffusion to Retouch on the Shoulder of Expertscited by 14
  20. 2024Osmosis: RGBD Diffusion Prior for Underwater Image Restorationcited by 13
  21. 2024Detail-Preserving Latent Diffusion for Stable Shadow Removalcited by 13
  22. 2024Taming Diffusion Models for Image Restoration: A Reviewcited by 12
  23. 2024Regional Attention for Shadow Removalcited by 12
  24. 2024Benchmarking Adversarial Robustness of Image Shadow Removal with Shadow-adaptive Attackscited by 10
  25. 2024Temporal Residual Guided Diffusion Framework for Event-Driven Video Reconstructioncited by 10
  26. 2024Diff-Shadow: Global-guided Diffusion Model for Shadow Removalcited by 10
  27. 2024Towards Image Ambient Lighting Normalizationcited by 9
  28. 2023JoReS-Diff: Joint Retinex and Semantic Priors in Diffusion Model for Low-light Image Enhancementcited by 8
  29. 2025Image Editing with Diffusion Models: A Surveycited by 8
  30. 2024ShadowMaskFormer: Mask Augmented Patch Embeddings for Shadow Removalcited by 7
  31. 2024Single-Image Shadow Removal Using Deep Learning: A Comprehensive Surveycited by 7
  32. 2024S3R-Net: A Single-Stage Approach to Self-Supervised Shadow Removalcited by 6
  33. 2024RSHazeDiff: A Unified Fourier-aware Diffusion Model for Remote Sensing Image Dehazingcited by 6
  34. 2024Unveiling Deep Shadows: A Survey and Benchmark on Image and Video Shadow Detection, Removal, and Generation in the Deep Learning Eracited by 6
  35. 2024Shadow Removal Refinement via Material-Consistent Shadow Edgescited by 5
  36. 2024MetaShadow: Object-Centered Shadow Detection, Removal, and Synthesiscited by 5
  37. 2025DenseSR: Image Shadow Removal as Dense Predictioncited by 5
  38. 2025UniSER: A Foundation Model for Unified Soft Effects Removalcited by 5

50 most cited of 102 All 102 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2023Refusion: Enabling Large-Size Realistic Image Restoration with Latent-Space Diffusion Models
  2. 2024Generative Portrait Shadow Removal
  3. 2023Generative Diffusion Prior for Unified Image Restoration and Enhancement
  4. 2025Exploiting Diffusion Prior for Real-World Image Dehazing with Unpaired Training
  5. 2025Residual Prior Diffusion: A Probabilistic Framework Integrating Coarse Latent Priors with Diffusion Models
  6. 2025UniLDiff: Unlocking the Power of Diffusion Priors for All-in-One Image Restoration
  7. 2023A Unified Conditional Framework for Diffusion-based Image Restoration
  8. 2026DiffRGD: An Inference-Time Diffusion Guidance Through Riemannian Gradient Descent