mlsanityData to 14 Oct 2026

ProPainter: Improving Propagation and Transformer for Video Inpainting

arXiv 2309.03897 ↗First posted 7 Sep 2023cs.CVcited by 197

Not on the map. Why?

Cited by 48 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Flow-based propagation and spatiotemporal Transformer are two mainstream mechanisms in video inpainting (VI). Despite the effectiveness of these components, they still suffer from some limitations that affect their performance. Previous propagation-based approaches are performed separately either in the image or feature domain. Global image propagation isolated from learning may cause spatial misalignment due to inaccurate optical flow. Moreover, memory or computational constraints limit the temporal range of feature propagation and video Transformer, preventing exploration of correspondence information from distant frames. To address these issues, we propose an improved framework, called ProPainter, which involves enhanced ProPagation and an efficient Transformer. Specifically, we introduce dual-domain propagation that combines the advantages of image and feature warping, exploiting global correspondences reliably. We also propose a mask-guided sparse video Transformer, which achieves high efficiency by discarding unnecessary and redundant tokens. With these components, ProPainter outperforms prior arts by a large margin of 1.46 dB in PSNR while maintaining appealing efficiency.

Key facts. Cited by 197 papers in the map (data through 14 Oct 2026). Builds on 36 of its 46 references that are in the map. First posted on arXiv 7 Sep 2023.

Builds on

36 of 46 references.

References · 36

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  3. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  4. 2017Deformable Convolutional Networkscited by 1,488
  5. 2021DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsificationcited by 864
  6. 2018Deformable ConvNets v2: More Deformable, Better Resultscited by 670
  7. 2018Video-to-Video Synthesiscited by 390
  8. 2022Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizationscited by 363
  9. 2021AdaViT: Adaptive Tokens for Efficient Vision Transformercited by 362
  10. 2021BasicVSR++: Improving Video Super-Resolution with Enhanced Propagation and Alignmentcited by 342
  11. 2020BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyondcited by 304
  12. 2018Learning Blind Video Temporal Consistencycited by 253
24 more
  1. 2018YouTube-VOS: Sequence-to-Sequence Video Object Segmentationcited by 232
  2. 2021AdaViT: Adaptive Vision Transformers for Efficient Image Recognitioncited by 228
  3. 2017UnFlow: Unsupervised Learning of Optical Flow with a Bidirectional Census Losscited by 202
  4. 2020Learning Joint Spatial-Temporal Transformations for Video Inpaintingcited by 163
  5. 2021Investigating Tradeoffs in Real-World Video Super-Resolutioncited by 134
  6. 2019Deep Flow-Guided Video Inpaintingcited by 116
  7. 2022Towards An End-to-End Framework for Flow-Guided Video Inpaintingcited by 113
  8. 2019Deep Video Inpaintingcited by 109
  9. 2019Spatio-Temporal Filter Adaptive Network for Video Deblurringcited by 106
  10. 2020Flow-edge Guided Video Completioncited by 104
  11. 2021FuseFormer: Fusing Fine-Grained Information in Transformers for Video Inpaintingcited by 95
  12. 2019Free-form Video Inpainting with 3D Gated Convolution and Temporal PatchGANcited by 87
  13. 2022Flow-Guided Transformer for Video Inpaintingcited by 59
  14. 2018Video Inpainting by Jointly Learning Temporal Structure and Spatial Detailscited by 58
  15. 2019Copy-and-Paste Networks for Deep Video Inpaintingcited by 58
  16. 2019Onion-Peel Networks for Deep Video Completioncited by 46
  17. 2020Short-Term and Long-Term Context Aggregation Network for Video Inpaintingcited by 35
  18. 2021Progressive Temporal Feature Alignment Network for Video Inpaintingcited by 35
  19. 2019An Internal Learning Approach to Video Inpaintingcited by 33
  20. 2021Occlusion-Aware Video Object Inpaintingcited by 30
  21. 2021Decoupled Spatial-Temporal Transformer for Video Inpaintingcited by 28
  22. 2021Internal Video Inpainting by Implicit Long-range Propagationcited by 19
  23. 2022On the Generalization of BasicVSR++ to Video Deblurring and Denoisingcited by 19
  24. 2022DeViT: Deformed Vision Transformers in Video Inpaintingcited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 6; 2024: 46; 2025: 73; 2026: 72.

197 papers.

Sort:

By area:

  1. 2025VACE: All-in-One Video Creation and Editingcited by 479
  2. 2023Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolutioncited by 132
  3. 2025VideoPainter: Any-length Video Inpainting and Editing with Plug-and-Play Context Controlcited by 84
  4. 2024MIMO: Controllable Character Video Synthesis with Spatial Decomposed Modelingcited by 64
  5. 2024DreamScene4D: Dynamic Multi-Object Scene Generation from Monocular Videoscited by 63
  6. 2025Follow-Your-Creation: Empowering 4D Creation through Video Inpaintingcited by 62
  7. 2023Dense Optical Tracking: Connecting the Dotscited by 61
  8. 2025Animate Anyone 2: High-Fidelity Character Image Animation with Environment Affordancecited by 61
  9. 2025Se~norita-2M: A High-Quality Instruction-based Dataset for General Video Editing by Video Specialistscited by 61
  10. 2024InFusion: Inpainting 3D Gaussians via Learning Depth Completion from Diffusion Priorcited by 56
  11. 2024IDOL: Instant Photorealistic 3D Human Creation from a Single Imagecited by 55
  12. 2026Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Modelscited by 50
38 more
  1. 2025MiniMax-Remover: Taming Bad Noise Helps Video Object Removalcited by 45
  2. 2024Generative Omnimatte: Learning to Decompose Video into Layerscited by 38
  3. 2025ROSE: Remove Objects with Side Effects in Videoscited by 38
  4. 2024StereoCrafter: Diffusion-based Generation of Long and High-fidelity Stereoscopic 3D from Monocular Videoscited by 37
  5. 2024Generative Video Propagationcited by 36
  6. 2025VideoShield: Regulating Diffusion-based Video Generation Models via Watermarkingcited by 34
  7. 2024Towards Language-Driven Video Inpainting via Multimodal Large Language Modelscited by 33
  8. 2024MVInpainter: Learning Multi-View Consistent Inpainting to Bridge 2D and 3D Editingcited by 30
  9. 2024V2A-Mark: Versatile Deep Visual-Audio Watermarking for Manipulation Localization and Copyright Protectioncited by 28
  10. 2024StableV2V: Stablizing Shape Consistency in Video-to-Video Editingcited by 26
  11. 2024VIVID-10M: A Dataset and Baseline for Versatile and Interactive Video Local Editingcited by 25
  12. 2025MimicDreamer: Aligning Human and Robot Demonstrations for Scalable VLA Trainingcited by 25
  13. 2025MatAnyone: Stable Video Matting with Consistent Memory Propagationcited by 24
  14. 2025RORem: Training a Robust Object Remover with Human-in-the-Loopcited by 23
  15. 2025ReBot: Scaling Robot Learning with Real-to-Sim-to-Real Robotic Video Synthesiscited by 22
  16. 20264RC: 4D Reconstruction via Conditional Querying Anytime and Anywherecited by 22
  17. 2024Video Diffusion Models are Strong Video Inpaintercited by 21
  18. 2024VASE: Object-Centric Appearance and Shape Manipulation of Real Videoscited by 20
  19. 2024SVG: 3D Stereoscopic Video Generation via Denoising Frame Matrixcited by 20
  20. 2023VIDiff: Translating Videos via Multi-Modal Instructions with Diffusion Modelscited by 19
  21. 2024Chat-Edit-3D: Interactive 3D Scene Editing via Text Promptscited by 19
  22. 2025OmniV2V: Versatile Video Generation and Editing via Dynamic Content Manipulationcited by 19
  23. 2025AirExo-2: Scaling up Generalizable Robotic Imitation Learning with Low-Cost Exoskeletonscited by 18
  24. 2024Blur-aware Spatio-temporal Sparse Transformer for Video Deblurringcited by 17
  25. 2024DriveEditor: A Unified 3D Information-Guided Framework for Controllable Object Editing in Driving Scenescited by 17
  26. 2025OmnimatteZero: Fast Training-free Omnimatte with Pre-trained Video Diffusion Modelscited by 16
  27. 2026NTIRE 2026 Challenge on Bitstream-Corrupted Video Restoration: Methods and Resultscited by 16
  28. 2026EffectErase: Joint Video Object Removal and Insertion for High-Quality Effect Erasingcited by 15
  29. 2025MTV-Inpaint: Multi-Task Long Video Inpaintingcited by 14
  30. 2025UltraVSR: Achieving Ultra-Realistic Video Super-Resolution with Efficient One-Step Diffusion Spacecited by 14
  31. 2025Precise Object and Effect Removal with Adaptive Target-Aware Attentioncited by 13
  32. 2023Cinematic Behavior Transfer via NeRF-based Differentiable Filmingcited by 12
  33. 2024VidPanos: Generative Panoramic Videos from Casual Panning Videoscited by 12
  34. 2024MovieCharacter: A Tuning-Free Framework for Controllable Character Video Synthesiscited by 12
  35. 2025Get In Video: Add Anything You Want to the Videocited by 12
  36. 2025UniLumos: Fast and Unified Image and Video Relighting with Physics-Plausible Feedbackcited by 12
  37. 2023Flow-Guided Diffusion for Video Inpaintingcited by 11
  38. 2025Vivid4D: Improving 4D Reconstruction from Monocular Video by Video Inpaintingcited by 11

50 most cited of 197 All 197 (JSON)