mlsanityData to 14 Oct 2026

Real-Time High-Resolution Background Matting

arXiv 2012.07810 ↗First posted 14 Dec 2020cs.CVcited by 173

Not on the map. Why?

Abstract (arXiv)

We introduce a real-time, high-resolution background replacement technique which operates at 30fps in 4K resolution, and 60fps for HD on a modern GPU. Our technique is based on background matting, where an additional frame of the background is captured and used in recovering the alpha matte and the foreground layer. The main challenge is to compute a high-quality alpha matte, preserving strand-level hair details, while processing high-resolution images in real-time. To achieve this goal, we employ two neural networks; a base network computes a low-resolution result which is refined by a second network operating at high-resolution on selective patches. We introduce two largescale video and image matting datasets: VideoMatte240K and PhotoMatte13K/85. Our approach yields higher quality results compared to the previous state-of-the-art in background matting, while simultaneously yielding a dramatic boost in both speed and resolution.

Key facts. Cited by 173 papers in the map (data through 14 Oct 2026). First posted on arXiv 14 Dec 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 20; 2022: 32; 2023: 33; 2024: 25; 2025: 39; 2026: 24.

173 papers.

Sort:

By area:

  1. 2023GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussianscited by 260
  2. 2022Tensor4D : Efficient Neural 4D Decomposition for High-fidelity Dynamic Reconstruction and Renderingcited by 217
  3. 2023NeRSemble: Multi-view Radiance Field Reconstruction of Human Headscited by 194
  4. 2023PAC-NeRF: Physics Augmented Continuum Neural Radiance Fields for Geometry-Agnostic System Identificationcited by 154
  5. 2021Robust High-Resolution Video Matting with Temporal Guidancecited by 137
  6. 2022Structured Local Radiance Fields for Human Avatar Modelingcited by 116
  7. 2023DNA-Rendering: A Diverse Neural Actor Repository for High-Fidelity Human-centric Renderingcited by 103
  8. 2023AvatarReX: Real-time Expressive Full-body Avatarscited by 84
  9. 2023HiFi4G: High-Fidelity Human Performance Rendering via Compact Gaussian Splattingcited by 70
  10. 2024SwitchLight: Co-design of Physics-driven Architecture and Pre-training Framework for Human Portrait Relightingcited by 67
  11. 2021Low-light Image Enhancement via Breaking Down the Darknesscited by 65
  12. 2021Privacy-Preserving Portrait Mattingcited by 60
38 more
  1. 2025FreeTimeGS: Free Gaussian Primitives at Anytime and Anywhere for Dynamic Scene Reconstructioncited by 54
  2. 2025Avat3r: Large Animatable Gaussian Reconstruction Model for High-fidelity 3D Head Avatarscited by 53
  3. 2024GIC: Gaussian-Informed Continuum for Physical Property Identification and Simulationcited by 49
  4. 2022MatteFormer: Transformer-Based Image Matting via Prior-Tokenscited by 48
  5. 2022NeuralHOFusion: Neural Volumetric Rendering under Human-object Interactionscited by 44
  6. 2023DiffusionAvatars: Deferred Diffusion for High-fidelity 3D Head Avatarscited by 42
  7. 2024Generative Omnimatte: Learning to Decompose Video into Layerscited by 38
  8. 2023HAvatar: High-fidelity Head Avatar via Facial Model Conditioned Neural Radiance Fieldcited by 37
  9. 2024GAF: Gaussian Avatar Reconstruction from Monocular Videos via Multi-view Diffusioncited by 36
  10. 2022Virtual Elastic Objectscited by 35
  11. 2022Reconstructing Hand-Held Objects from Monocular Videocited by 35
  12. 2023HHAvatar: Gaussian Head Avatar with Dynamic Hairscited by 35
  13. 2023OpenIllumination: A Multi-Illumination Dataset for Inverse Rendering Evaluation on Real Objectscited by 31
  14. 2023AvatarStudio: Text-driven Editing of 3D Dynamic Human Head Avatarscited by 30
  15. 2024Multi-view Aggregation Network for Dichotomous Image Segmentationcited by 28
  16. 2021Attention-guided Temporally Coherent Video Object Mattingcited by 25
  17. 2022Rethinking Portrait Matting with Privacy Preservingcited by 25
  18. 2022Human Instance Matting via Mutual Guidance and Multi-Instance Refinementcited by 25
  19. 2025Human Motion Video Generation: A Surveycited by 25
  20. 2025MatAnyone: Stable Video Matting with Consistent Memory Propagationcited by 24
  21. 2025TransAnimate: Taming Layer Diffusion to Generate RGBA Videocited by 24
  22. 2022PP-Matting: High-Accuracy Natural Image Mattingcited by 23
  23. 2022Boosting Robustness of Image Matting with Context Assembling and Strong Data Augmentationcited by 22
  24. 2022Referring Image Mattingcited by 22
  25. 2022VMFormer: End-to-End Video Matting with Transformercited by 20
  26. 2025Vid2Sim: Generalizable, Video-based Reconstruction of Appearance, Geometry and Physics for Mesh-free Simulationcited by 20
  27. 2021Blind Image Decompositioncited by 18
  28. 2024PKU-DyMVHumans: A Multi-View Video Benchmark for High-Fidelity Dynamic Human Modelingcited by 18
  29. 2024MaGGIe: Masked Guided Gradual Human Instance Mattingcited by 18
  30. 2022MPIB: An MPI-Based Bokeh Rendering Framework for Realistic Partial Occlusion Effectscited by 17
  31. 2023Holoported Characters: Real-time Free-viewpoint Rendering of Humans from Sparse RGB Camerascited by 17
  32. 2025Diffuman4D: 4D Consistent Human View Synthesis from Sparse-View Videos with Spatio-Temporal Diffusion Modelscited by 17
  33. 2024Improving Physics-Augmented Continuum Neural Radiance Field-Based Geometry-Agnostic System Identification with Lagrangian Particle Optimizationcited by 16
  34. 2021Image2Lego: Customized LEGO Set Generation from Imagescited by 14
  35. 2021VaxNeRF: Revisiting the Classic for Voxel-Accelerated Neural Radiance Fieldcited by 14
  36. 2022Multi-sensor large-scale dataset for multi-view 3D reconstructioncited by 14
  37. 2023OmnimatteRF: Robust Omnimatte with 3D Background Modelingcited by 14
  38. 2026Video Generation Models are General-Purpose Vision Learnerscited by 14

50 most cited of 173 All 173 (JSON)