mlsanityData to 14 Oct 2026

SpatialTrackerV2: 3D Point Tracking Made Easy

arXiv 2507.12462 ↗First posted 16 Jul 2025cs.CVcited by 87

Not on the map. Why?

Abstract (arXiv)

We present SpatialTrackerV2, a feed-forward 3D point tracking method for monocular videos. Going beyond modular pipelines built on off-the-shelf components for 3D tracking, our approach unifies the intrinsic connections between point tracking, monocular depth, and camera pose estimation into a high-performing and feedforward 3D point tracker. It decomposes world-space 3D motion into scene geometry, camera ego-motion, and pixel-wise object motion, with a fully differentiable and end-to-end architecture, allowing scalable training across a wide range of datasets, including synthetic sequences, posed RGB-D videos, and unlabeled in-the-wild footage. By learning geometry and motion jointly from such heterogeneous data, SpatialTrackerV2 outperforms existing 3D tracking methods by 30%, and matches the accuracy of leading dynamic 3D reconstruction approaches while running 50× faster.

Key facts. Cited by 87 papers in the map (data through 14 Oct 2026). Builds on 75 of its 97 references that are in the map. First posted on arXiv 16 Jul 2025.

Builds on

75 of 97 references.

References · 75

  1. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  2. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  3. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  4. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  5. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,567
  6. 2021Ego4D: Around the World in 3,000 Hours of Egocentric Videocited by 1,556
  7. 2025VGGT: Visual Geometry Grounded Transformercited by 1,539
  8. 2024Depth Anything V2cited by 1,422
  9. 2024Depth Anything: Unleashing the Power of Large-Scale Unlabeled Datacited by 1,360
  10. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,274
  11. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 881
  12. 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 775
38 more
  1. 2018Deep Ordinal Regression Network for Monocular Depth Estimationcited by 723
  2. 2023ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depthcited by 657
  3. 2021Common Objects in 3D: Large-Scale Learning and Evaluation of Real-life 3D Category Reconstructioncited by 652
  4. 2025Continuous 3D Perception Model with Persistent Statecited by 548
  5. 2023DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Visioncited by 537
  6. 2023CoTracker: It is Better to Track Togethercited by 522
  7. 2020TartanAir: A Dataset to Push the Limits of Visual SLAMcited by 506
  8. 2024MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motioncited by 442
  9. 2024Metric3Dv2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimationcited by 395
  10. 2024UniDepth: Universal Monocular Metric Depth Estimationcited by 382
  11. 2022Kubric: A scalable dataset generatorcited by 381
  12. 2022HOI4D: A 4D Egocentric Dataset for Category-Level Human-Object Interactioncited by 327
  13. 2023OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic Perception, Reconstruction and Generationcited by 326
  14. 2024MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervisioncited by 299
  15. 2023TAPIR: Tracking Any Point with per-frame Initialization and temporal Refinementcited by 287
  16. 2023PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Trackingcited by 276
  17. 2022TAP-Vid: A Benchmark for Tracking Any Point in a Videocited by 263
  18. 2024Shape of Motion: 4D Reconstruction from a Single Videocited by 250
  19. 2017SSD-6D: Making RGB-based 3D detection and 6D pose estimation great againcited by 243
  20. 2023BEDLAM: A Synthetic Dataset of Bodies Exhibiting Detailed Lifelike Animated Motioncited by 236
  21. 2024MegaSaM: Accurate, Fast, and Robust Structure and Motion from Casual Dynamic Videoscited by 228
  22. 2023Visual Geometry Grounded Deep Structure From Motioncited by 227
  23. 2022Particle Video Revisited: Tracking Through Occlusions Using Point Trajectoriescited by 217
  24. 2025Video Depth Anything: Consistent Depth Estimation for Super-Long Videoscited by 207
  25. 2024DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videoscited by 204
  26. 2023Tracking Everything Everywhere All at Oncecited by 193
  27. 2020Robust Consistent Video Depth Estimationcited by 188
  28. 2025Diffusion as Shader: 3D-aware Video Diffusion for Versatile Video Generation Controlcited by 185
  29. 2019ReFusion: 3D Reconstruction in Dynamic Environments for RGB-D Cameras Exploiting Residualscited by 181
  30. 2024SpatialTracker: Tracking Any 2D Pixels in 3D Spacecited by 181
  31. 2024MoSca: Dynamic Gaussian Fusion from Casual Videos via 4D Motion Scaffoldscited by 179
  32. 2023MiDaS v3.1 -- A Model Zoo for Robust Monocular Relative Depth Estimationcited by 174
  33. 2023Spring: A High-Resolution High-Detail Dataset and Benchmark for Scene Flow, Optical Flow and Stereocited by 172
  34. 2017A Survey of Structure from Motioncited by 158
  35. 2022NeW CRFs: Neural Window Fully-connected CRFs for Monocular Depth Estimationcited by 153
  36. 2023DynamicStereo: Consistent Dynamic Depth from Stereo Videoscited by 148
  37. 2023PoseDiffusion: Solving Pose Estimation via Diffusion-aided Bundle Adjustmentcited by 147
  38. 2024Cameras as Rays: Pose Estimation via Ray Diffusioncited by 131

50 most cited of 75 All 75 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 32; 2026: 55.

87 papers.

Sort:

By area:

  1. 2025ViPE: Video Pose Engine for 3D Geometric Perceptioncited by 170
  2. 2026LingBot-Map: Geometric Context Transformer for Streaming 3D Reconstructioncited by 37
  3. 2025Any4D: Unified Feed-Forward Metric 4D Reconstructioncited by 35
  4. 2025Trace Anything: Representing Any Video in 4D via Trajectory Fieldscited by 33
  5. 2025What Is The Best 3D Scene Representation for Robotics? From Geometric to Foundation Modelscited by 29
  6. 2025IGGT: Instance-Grounded Geometry Transformer for Semantic 3D Reconstructioncited by 28
  7. 2025VGGT4D: Mining Motion Cues in Visual Geometry Transformers for 4D Scene Reconstructioncited by 28
  8. 2026Masked Depth Modeling for Spatial Perceptioncited by 28
  9. 2026VGGT-Omegacited by 26
  10. 2025Dream2Flow: Bridging Video Generation and Open-World Manipulation with 3D Object Flowcited by 25
  11. 2025DSI-Bench: A Benchmark for Dynamic Spatial Intelligencecited by 23
  12. 2025Generative Video Motion Editing with 3D Point Trackscited by 23
38 more
  1. 20264RC: 4D Reconstruction via Conditional Querying Anytime and Anywherecited by 22
  2. 2026EgoWAM: World Action Models Beyond Pixels with In-the-Wild Egocentric Human Datacited by 20
  3. 2025Multi-View 3D Point Trackingcited by 17
  4. 2025VideoArtGS: Building Digital Twins of Articulated Objects from Monocular Videocited by 15
  5. 2025TraceGen: World Modeling in 3D Trace Space Enables Learning from Cross-Embodiment Videoscited by 15
  6. 2026Motion 3-to-4: 3D Motion Reconstruction for 4D Synthesiscited by 15
  7. 2025Reconstructing 4D Spatial Intelligence: A Surveycited by 14
  8. 2026Do as I Do: Dexterous Manipulation Data from Everyday Human Videoscited by 13
  9. 2025SpatialTree: How Spatial Abilities Branch Out in MLLMscited by 12
  10. 2026Kinema4D: Kinematic 4D World Modeling for Spatiotemporal Embodied Simulationcited by 12
  11. 2025SpaceTimePilot: Generative Rendering of Dynamic Scenes Across Space and Timecited by 11
  12. 2025DexMan: Learning Bimanual Dexterous Manipulation from Human and Generated Videoscited by 10
  13. 2025UniSplat: Unified Spatio-Temporal Fusion via 3D Latent Scaffolds for Dynamic Driving Scene Reconstructioncited by 10
  14. 2025EcoSplat: Efficiency-controllable Feed-forward 3D Gaussian Splatting from Multi-view Imagescited by 10
  15. 2026SymphoMotion: Joint Control of Camera Motion and Object Dynamics for Coherent Video Generationcited by 10
  16. 2026TC-IDM: Grounding Video Generation for Executable Zero-shot Robot Motioncited by 8
  17. 2026Flow4R: Unifying 4D Reconstruction and Tracking with Scene Flowcited by 8
  18. 2026WHOLE: World-Grounded Hand-Object Lifted from Egocentric Videoscited by 8
  19. 2026Pri4R: Learning World Dynamics for Vision-Language-Action Models with Privileged 4D Representationcited by 8
  20. 2025IC-World: In-Context Generation for Shared World Modelingcited by 6
  21. 2025SeeU: Seeing the Unseen World via 4D Dynamics-aware Generationcited by 5
  22. 2026TrackCraft3R: Repurposing Video Diffusion Transformers for Dense 3D Trackingcited by 5
  23. 2026DynaFLIP: Rethinking Robotics Perception via Tri-Modal-Dynamics Guided Representationcited by 5
  24. 2026World Tracing: Generative Pixel-Aligned Geometry Beyond the Visiblecited by 5
  25. 2026MolmoMotion: Forecasting Point Trajectories in 3D with Language Instructioncited by 5
  26. 2025Advances in 4D Representation: Geometry, Motion, and Interactioncited by 4
  27. 2025Seeing without Pixels: Perception from Camera Trajectoriescited by 4
  28. 2025Look Around and Pay Attention: Multi-camera Point Tracking Reimagined with Transformerscited by 4
  29. 2026TrajVG: 3D Trajectory-Coupled Visual Geometry Learningcited by 4
  30. 2026ArtHOI: Taming Foundation Models for Monocular 4D Reconstruction of Hand-Articulated-Object Interactionscited by 4
  31. 2025Probing and Leveraging Video Diffusion Transformer Features for Robust Point Trackingcited by 3
  32. 2026FreeArtGS: Articulated Gaussian Splatting Under Free-moving Scenariocited by 3
  33. 2025What If : Understanding Motion Through Sparse Interactionscited by 2
  34. 2025Target-Bench: Can Video World Models Achieve Mapless Path Planning with Semantic Targets?cited by 2
  35. 2025Geometry-Aware Single-Image 4D Synthesis via Dense Trajectory Generationcited by 2
  36. 2026Robust 4D Visual Geometry Transformer with Uncertainty-Aware Priorscited by 2
  37. 2026Imagine2Real: Towards Zero-shot Humanoid-Object Interaction via Video Generative Priorscited by 2
  38. 2026ELAN4D: Embodiment-Centric 4D Supervision for Vision-Language-Action Models via Plug-and-Play Adaptationcited by 2

50 most cited of 87 All 87 (JSON)