mlsanityData to 7 Oct 2026

MultiPhys: Multi-Person Physics-aware 3D Motion Estimation

arXiv 2404.11987 ↗First posted 18 Apr 2024cs.CVcited by 18

Not on the map. Why?

Abstract (arXiv)

We introduce MultiPhys, a method designed for recovering multi-person motion from monocular videos. Our focus lies in capturing coherent spatial placement between pairs of individuals across varying degrees of engagement. MultiPhys, being physically aware, exhibits robustness to jittering and occlusions, and effectively eliminates penetration issues between the two individuals. We devise a pipeline in which the motion estimated by a kinematic-based method is fed into a physics simulator in an autoregressive manner. We introduce distinct components that enable our model to harness the simulator's properties without compromising the accuracy of the kinematic estimates. This results in final motion estimates that are both kinematically coherent and physically compliant. Extensive evaluations on three challenging datasets characterized by substantial inter-person interaction show that our method significantly reduces errors associated with penetration and foot skating, while performing competitively with the state-of-the-art on motion accuracy and smoothness. Results and code can be found on our project page (http://www.iri.upc.edu/people/nugrinovic/multiphys/).

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 33 of its 49 references that are in the map. First posted on arXiv 18 Apr 2024.

Builds on

33 of 49 references.

References · 33

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2019Expressive Body Capture: 3D Hands, Face, and Body from a Single Imagecited by 1,600
  3. 2019AMASS: Archive of Motion Capture as Surface Shapescited by 1,208
  4. 2017End-to-end Recovery of Human Shape and Posecited by 883
  5. 2019VIBE: Video Inference for Human Body Pose and Shape Estimationcited by 615
  6. 2023Humans in 4D: Reconstructing and Tracking Humans with Transformerscited by 344
  7. 2021HuMoR: 3D Human Motion Model for Robust Pose Estimationcited by 319
  8. 2018Learning 3D Human Dynamics from Videocited by 315
  9. 2020Monocular, One-stage, Regression of Multiple 3D Peoplecited by 216
  10. 2017Single-Shot Multi-Person 3D Pose Estimation From Monocular RGBcited by 195
  11. 2020Beyond Static Features for Temporally Consistent 3D Human Pose and Shape from a Videocited by 162
  12. 2021GLAMR: Global Occlusion-Aware Human Mesh Recovery with Dynamic Camerascited by 151
21 more
  1. 2021Putting People in their Place: Monocular Regression of 3D People in Depthcited by 141
  2. 2023Decoupling Human and Camera Motion from Videos in the Wildcited by 140
  3. 2020Coherent Reconstruction of Multiple Humans from a Single Imagecited by 136
  4. 2021SimPoE: Simulated Character Control for 3D Human Pose Estimationcited by 111
  5. 2022AvatarPoser: Articulated Full-Body Pose Tracking from Sparse Motion Sensingcited by 107
  6. 2021Dynamics-Regulated Kinematic Policy for Egocentric Pose Estimationcited by 101
  7. 20203D Human Motion Estimation via Motion Compression and Refinementcited by 98
  8. 2023TRACE: 5D Temporal Regression of Avatars with Dynamic Cameras in 3D Environmentscited by 89
  9. 2021Multi-Person Extreme Motion Predictioncited by 74
  10. 2022QuestSim: Human Motion Tracking from Sparse Sensors with Simulated Avatarscited by 74
  11. 2021Neural Monocular 3D Human Motion Capture with Physical Awarenesscited by 70
  12. 2020Residual Force Control for Agile Human Behavior Imitation and Extended Motion Synthesiscited by 63
  13. 2020Human Mesh Recovery from Multiple Shotscited by 52
  14. 2022Occluded Human Mesh Recoverycited by 52
  15. 2023Generative Proxemics: A Prior for 3D Social Interaction from Imagescited by 43
  16. 2023Realistic Full-Body Tracking from Sparse Observations via Joint-Level Modelingcited by 41
  17. 2022D&D: Learning Human Dynamics from Dynamic Cameracited by 39
  18. 2022Differentiable Dynamics for Articulated 3d Human Motion Reconstructioncited by 38
  19. 2022Trajectory Optimization for Physics-Based Reconstruction of 3d Human Pose from Monocular Videocited by 37
  20. 2023QuestEnvSim: Environment-Aware Simulated Motion Tracking from Sparse Sensorscited by 26
  21. 2021Body Size and Depth Disambiguation in Multi-Person Reconstruction from Single Imagescited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 9; 2026: 9.

18 papers.

Sort:

  1. 2025Visual Imitation Enables Contextual Humanoid Controlcited by 57
  2. 2025Reconstructing Humans with a Biomechanically Accurate Skeletoncited by 22
  3. 2025Reconstructing 4D Spatial Intelligence: A Surveycited by 13
  4. 2026RAM: Recover Any 3D Human Motion in-the-Wildcited by 11
  5. 2025Ponimator: Unfolding Interactive Pose for Versatile Human-human Interaction Animationcited by 7
  6. 2025PhysiInter: Integrating Physical Mapping for High-Fidelity Human Interaction Generationcited by 4
  7. 2025Reconstructing Close Human Interaction with Appearance and Proxemics Reasoningcited by 4
  8. 2025PP-Motion: Physical-Perceptual Fidelity Evaluation for Human Motion Generationcited by 4
  9. 2025Group Inertial Poser: Multi-Person Pose and Global Translation from Sparse Inertial Sensors and Ultra-Wideband Rangingcited by 3
  10. 2026HumanScore: Benchmarking Human Motions in Generated Videoscited by 2
  11. 2025Multi-person Physics-based Pose Estimation for Combat Sportscited by 1
  12. 2026SocialMirror: Reconstructing 3D Human Interaction Behaviors from Monocular Videos with Semantic and Geometric Guidancecited by 0
6 more
  1. 2026Recovering Physically Plausible Human-Object Interactions from Monocular Videoscited by 0
  2. 2026TaskNPoint: How to Teach Your Humanoid to Hit a Backhand in Minutescited by 0
  3. 2026GNOCHI: Generative Neural mOdel for Close Human-Human Interactionscited by 0
  4. 2026EgoExoMoCap: Distributed Ego-Exo Human Motion Capturecited by 0
  5. 2026Inter-X++: A Comprehensive Benchmark for Multimodal Human-Human Interaction Analysiscited by 0
  6. 2026WeLike2Party! In-Context Motion Transfer for Multi-Human Image Animationcited by 0