mlsanityData to 14 Oct 2026

MFT: Long-Term Tracking of Every Pixel

arXiv 2305.12998 ↗First posted 22 May 2023cs.CVcited by 62

Not on the map. Why?

Abstract (arXiv)

We propose MFT -- Multi-Flow dense Tracker -- a novel method for dense, pixel-level, long-term tracking. The approach exploits optical flows estimated not only between consecutive frames, but also for pairs of frames at logarithmically spaced intervals. It selects the most reliable sequence of flows on the basis of estimates of its geometric accuracy and the probability of occlusion, both provided by a pre-trained CNN. We show that MFT achieves competitive performance on the TAP-Vid benchmark, outperforming baselines by a significant margin, and tracking densely orders of magnitude faster than the state-of-the-art point-tracking methods. The method is insensitive to medium-length occlusions and it is robustified by estimating flow with respect to the reference frame, which reduces drift.

Key facts. Cited by 62 papers in the map (data through 14 Oct 2026). Builds on 37 of its 73 references that are in the map. First posted on arXiv 22 May 2023.

Builds on

37 of 73 references.

References · 37

  1. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,270
  2. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  3. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,446
  4. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,059
  5. 2017PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volumecited by 968
  6. 2017The 2017 DAVIS Challenge on Video Object Segmentationcited by 957
  7. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 881
  8. 2016FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networkscited by 784
  9. 2018YouTube-VOS: A Large-Scale Video Object Segmentation Benchmarkcited by 383
  10. 2022Kubric: A scalable dataset generatorcited by 381
  11. 2019A Short Note on the Kinetics-700 Human Action Datasetcited by 310
  12. 2018ATOM: Accurate Tracking by Overlap Maximizationcited by 285
25 more
  1. 2022FlowFormer: A Transformer Architecture for Optical Flowcited by 275
  2. 2019Learning Correspondence from the Cycle-Consistency of Timecited by 269
  3. 2022TAP-Vid: A Benchmark for Tracking Any Point in a Videocited by 263
  4. 2022Particle Video Revisited: Tracking Through Occlusions Using Point Trajectoriescited by 217
  5. 2021BANMo: Building Animatable 3D Neural Models from Many Casual Videoscited by 197
  6. 2020BOP Challenge 2020 on 6D Object Localizationcited by 193
  7. 2019DPOD: 6D Pose Object Detector and Refinercited by 189
  8. 2021COTR: Correspondence Transformer for Matching Across Imagescited by 183
  9. 2018Bounding Box Regression with Uncertainty for Accurate Object Detectioncited by 143
  10. 2018Uncertainty Estimates and Multi-Hypotheses Networks for Optical Flowcited by 135
  11. 2019Iterative Residual Refinement for Joint Optical Flow and Occlusion Estimationcited by 131
  12. 2020MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Maskcited by 117
  13. 2021LASR: Learning Articulated Shape Reconstruction from a Monocular Videocited by 116
  14. 2018Creatures great and SMAL: Recovering the shape and motion of animals from videocited by 84
  15. 2020Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loopcited by 81
  16. 2018Occlusions, Motion and Depth Boundaries with a Generic Network for Disparity, Optical Flow or Scene Flow Estimationcited by 69
  17. 2020Online Adaptation for Consistent Mesh Reconstruction in the Wildcited by 51
  18. 2018A Fusion Approach for Multi-Frame Optical Flow Estimationcited by 37
  19. 2022Refign: Align and Refine for Adaptation of Semantic Segmentation to Adverse Conditionscited by 31
  20. 2017ProbFlow: Joint Optical Flow and Uncertainty Estimationcited by 24
  21. 2021Homography Decomposition Networks for Planar Object Trackingcited by 9
  22. 2023Planar Object Tracking via Weighted Optical Flowcited by 7
  23. 2022DirectTracker: 3D Multi-Object Tracking Using Direct Image Alignment and Photometric Bundle Adjustmentcited by 6
  24. 2022NeuralMarker: A Framework for Learning General Marker Correspondencecited by 4
  25. 2016Drift Robust Non-rigid Optical Flow Enhancement for Long Sequencescited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 6; 2024: 23; 2025: 22; 2026: 11.

62 papers.

Sort:

By area:

  1. 2023CoTracker: It is Better to Track Togethercited by 522
  2. 2024Shape of Motion: 4D Reconstruction from a Single Videocited by 250
  3. 2023Tracking Everything Everywhere All at Oncecited by 193
  4. 2024SpatialTracker: Tracking Any 2D Pixels in 3D Spacecited by 181
  5. 2024BootsTAP: Bootstrapped Training for Tracking-Any-Pointcited by 120
  6. 2024Local All-Pair Correspondence for Point Trackingcited by 80
  7. 2024DELTA: Dense Efficient Long-range 3D Tracking for any videocited by 65
  8. 2024DINO-Tracker: Taming DINO for Self-Supervised Point Tracking in a Single Videocited by 64
  9. 2023Dense Optical Tracking: Connecting the Dotscited by 61
  10. 2025AllTracker: Efficient Dense Point Tracking at High Resolutioncited by 58
  11. 2024TAPTR: Tracking Any Point with Transformers as Detectioncited by 45
  12. 2023Tracking and Mapping in Medical Computer Vision: A Reviewcited by 36
38 more
  1. 2024Splatter a Video: Video Gaussian Representation for Versatile Processingcited by 31
  2. 2024SceneTracker: Long-term Scene Flow Estimation Networkcited by 26
  3. 2024TAPTRv2: Attention-based Position Update Improves Tracking Any Pointcited by 18
  4. 2024Autogenic Language Embedding for Coherent Point Trackingcited by 17
  5. 2023Surgical Tattoos in Infrared: A Dataset for Quantifying Tissue Tracking and Mappingcited by 15
  6. 2025DGS-LRM: Real-Time Deformable 3D Gaussian Reconstruction From Monocular Videoscited by 15
  7. 2024Track Everything Everywhere Fast and Robustlycited by 14
  8. 2024Trajectory-aligned Space-time Tokens for Few-shot Action Recognitioncited by 10
  9. 2024Self-Supervised Any-Point Tracking by Contrastive Random Walkscited by 10
  10. 2025DELTAv2: Accelerating Dense 3D Trackingcited by 10
  11. 2024TAPTRv3: Spatial and Temporal Context Foster Robust Tracking of Any Point in Long Videocited by 8
  12. 2025C4D: 4D Made from 3D through Dual Correspondencescited by 8
  13. 2023SplatFlow: Learning Multi-frame Optical Flow via Splattingcited by 7
  14. 2024MFTIQ: Multi-Flow Tracker with Independent Matching Quality Estimationcited by 7
  15. 2025Point Tracking in Surgery--The 2024 Surgical Tattoos in Infrared (STIR) Challengecited by 7
  16. 2025SciVid: Cross-Domain Evaluation of Video Models in Scientific Applicationscited by 6
  17. 2024Tracking Everything in Robotic-Assisted Surgerycited by 5
  18. 2025LiteTracker: Leveraging Temporal Causality for Accurate Low-latency Tissue Trackingcited by 5
  19. 2025Point Prompting: Counterfactual Tracking with Video Diffusion Modelscited by 5
  20. 2026Video Understanding: From Geometry and Semantics to Unified Modelscited by 5
  21. 2024A-MFST: Adaptive Multi-Flow Sparse Tracker for Real-Time Tissue Tracking Under Occlusioncited by 4
  22. 2024Ada-Tracker: Soft Tissue Tracking via Inter-Frame and Adaptive-Template Matchingcited by 3
  23. 2025ProTracker: Probabilistic Integration for Robust and Accurate Point Trackingcited by 3
  24. 2025Online Dense Point Tracking with Streaming Memorycited by 3
  25. 2025AnthroTAP: Learning Point Tracking with Real-World Motioncited by 3
  26. 2025Track-On2: Enhancing Online Point Tracking with Memorycited by 3
  27. 2025BleedOrigin: Dynamic Bleeding Source Localization in Endoscopic Submucosal Dissection via Dual-Stage Detection and Trackingcited by 2
  28. 2025Vectorized Video Representation with Easy Editing via Hierarchical Spatio-Temporally Consistent Proxy Embeddingcited by 2
  29. 2026VidMap: Exploiting Temporal Structure for Video-Based Structure-from-Motioncited by 2
  30. 2024Dense Matchers for Dense Trackingcited by 1
  31. 2024Decomposition Betters Tracking Everything Everywherecited by 1
  32. 2025Endo-TTAP: Robust Endoscopic Tissue Tracking via Multi-Facet Guided Attention and Hybrid Flow-point Supervisioncited by 1
  33. 2026Moaw: Unleashing Motion Awareness for Video Diffusion Modelscited by 1
  34. 20264D Monocular Surgical Reconstruction under Arbitrary Camera Motionscited by 1
  35. 2026Triangular Consistency as a Universal Constraint for Learning Optical Flowcited by 1
  36. 2024Leveraging Object Priors for Point Trackingcited by 0
  37. 2024Solution for Point Tracking Task of ECCV 2nd Perception Test Challenge 2024cited by 0
  38. 2025GIFT: Generated Indoor video frames for Texture-less point trackingcited by 0

50 most cited of 62 All 62 (JSON)