mlsanityData to 14 Oct 2026

RAFT: Recurrent All-Pairs Field Transforms for Optical Flow

arXiv 2003.12039 ↗First posted 26 Mar 2020cs.CVcited by 2,415

Not on the map. Why?

Cited by 281 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We introduce Recurrent All-Pairs Field Transforms (RAFT), a new deep network architecture for optical flow. RAFT extracts per-pixel features, builds multi-scale 4D correlation volumes for all pairs of pixels, and iteratively updates a flow field through a recurrent unit that performs lookups on the correlation volumes. RAFT achieves state-of-the-art performance. On KITTI, RAFT achieves an F1-all error of 5.10%, a 16% error reduction from the best published result (6.10%). On Sintel (final pass), RAFT obtains an end-point-error of 2.855 pixels, a 30% error reduction from the best published result (4.098 pixels). In addition, RAFT has strong cross-dataset generalization as well as high efficiency in inference time, training speed, and parameter count. Code is available at https://github.com/princeton-vl/RAFT.

Key facts. Cited by 2,415 papers in the map (data through 14 Oct 2026). First posted on arXiv 26 Mar 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 147; 2022: 283; 2023: 399; 2024: 536; 2025: 583; 2026: 467.

2,415 papers.

Sort:

By area:

  1. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasetscited by 2,581
  2. 2023VBench: Comprehensive Benchmark Suite for Video Generative Modelscited by 1,822
  3. 2021Is Space-Time Attention All You Need for Video Understanding?cited by 1,620
  4. 2024Grounding Image Matching in 3D with MASt3Rcited by 983
  5. 2024Emu3: Next-Token Prediction is All You Needcited by 781
  6. 2021DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Camerascited by 687
  7. 2021Perceiver IO: A General Architecture for Structured Inputs & Outputscited by 629
  8. 2023CoTracker: It is Better to Track Togethercited by 522
  9. 2025VACE: All-in-One Video Creation and Editingcited by 479
  10. 2023VideoComposer: Compositional Video Synthesis with Motion Controllabilitycited by 464
  11. 2024CameraCtrl: Enabling Camera Control for Text-to-Video Generationcited by 460
  12. 2023TokenFlow: Consistent Diffusion Features for Consistent Video Editingcited by 442
38 more
  1. 2025Step1X-Edit: A Practical Framework for General Image Editingcited by 429
  2. 2021Dynamic View Synthesis from Dynamic Monocular Videocited by 420
  3. 2024Metric3Dv2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimationcited by 395
  4. 2022Kubric: A scalable dataset generatorcited by 381
  5. 2021RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matchingcited by 353
  6. 2024OpenVid-1M: A Large-Scale High-Quality Dataset for Text-to-video Generationcited by 341
  7. 2023EvalCrafter: Benchmarking and Evaluating Large Video Generation Modelscited by 340
  8. 20243D-VLA: A 3D Vision-Language-Action Generative World Modelcited by 328
  9. 2021GMFlow: Learning Optical Flow via Global Matchingcited by 327
  10. 2023Pix2Video: Video Editing using Image Diffusioncited by 320
  11. 2025VBench-2.0: Advancing Video Generation Benchmark Suite for Intrinsic Faithfulnesscited by 315
  12. 2023Human Preference Score: Better Aligning Text-to-Image Models with Human Preferencecited by 287
  13. 2023TAPIR: Tracking Any Point with per-frame Initialization and temporal Refinementcited by 287
  14. 20243D Reconstruction with Spatial Memorycited by 277
  15. 2023PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Trackingcited by 276
  16. 2022FlowFormer: A Transformer Architecture for Optical Flowcited by 275
  17. 2022Unifying Flow, Stereo and Depth Estimationcited by 275
  18. 2022TAP-Vid: A Benchmark for Tracking Any Point in a Videocited by 263
  19. 2024Shape of Motion: 4D Reconstruction from a Single Videocited by 250
  20. 2022Unsupervised Semantic Segmentation by Distilling Feature Correspondencescited by 241
  21. 2024VideoPhy: Evaluating Physical Commonsense for Video Generationcited by 228
  22. 2024MegaSaM: Accurate, Fast, and Robust Structure and Motion from Casual Dynamic Videoscited by 228
  23. 2023Visual Geometry Grounded Deep Structure From Motioncited by 227
  24. 2021Learning to Estimate Hidden Motions with Global Motion Aggregationcited by 224
  25. 2021Keeping Your Eye on the Ball: Trajectory Attention in Video Transformerscited by 224
  26. 2022DynIBaR: Neural Dynamic Image-Based Renderingcited by 224
  27. 2023Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifoldcited by 219
  28. 2022Particle Video Revisited: Tracking Through Occlusions Using Point Trajectoriescited by 217
  29. 2024Motion-I2V: Consistent and Controllable Image-to-Video Generation with Explicit Motion Modelingcited by 211
  30. 2025Video Depth Anything: Consistent Depth Estimation for Super-Long Videoscited by 207
  31. 2024MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priorscited by 205
  32. 2021PIRenderer: Controllable Portrait Image Generation via Semantic Neural Renderingcited by 203
  33. 2022Monocular Dynamic View Synthesis: A Reality Checkcited by 200
  34. 2022Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlationcited by 197
  35. 2023ProPainter: Improving Propagation and Transformer for Video Inpaintingcited by 197
  36. 2023Tracking Anything with Decoupled Video Segmentationcited by 197
  37. 2023Robust Dynamic Radiance Fieldscited by 194
  38. 2023Tracking Everything Everywhere All at Oncecited by 193

50 most cited of 2,415 All 2,415 (JSON)