Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
143 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

GroundLink: A Dataset Unifying Human Body Movement and Ground Reaction Dynamics (2310.03930v1)

Published 5 Oct 2023 in cs.GR

Abstract: The physical plausibility of human motions is vital to various applications in fields including but not limited to graphics, animation, robotics, vision, biomechanics, and sports science. While fully simulating human motions with physics is an extreme challenge, we hypothesize that we can treat this complexity as a black box in a data-driven manner if we focus on the ground contact, and have sufficient observations of physics and human activities in the real world. To prove our hypothesis, we present GroundLink, a unified dataset comprised of captured ground reaction force (GRF) and center of pressure (CoP) synchronized to standard kinematic motion captures. GRF and CoP of GroundLink are not simulated but captured at high temporal resolution using force platforms embedded in the ground for uncompromising measurement accuracy. This dataset contains 368 processed motion trials (~1.59M recorded frames) with 19 different movements including locomotion and weight-shifting actions such as tennis swings to signify the importance of capturing physics paired with kinematics. GroundLinkNet, our benchmark neural network model trained with GroundLink, supports our hypothesis by predicting GRFs and CoPs accurately and plausibly on unseen motions from various sources. The dataset, code, and benchmark models are made public for further research on various downstream tasks leveraging the rich physics information at https://csr.bu.edu/groundlink/.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (66)
  1. OSU ACCAD. 2001. ACCAD MoCap System and Data. https://accad.osu.edu/research/motion-lab/mocap-system-and-data Accessed: 2023-05-01.
  2. On Learning Symmetric Locomotion. In Proc. ACM SIGGRAPH Motion, Interaction, and Games (MIG 2019).
  3. These legs were made for propulsion: advancing the diagnosis and treatment of post-stroke propulsion deficits. Journal of NeuroEngineering and Rehabilitation 17, 1 (2020), 139. https://doi.org/10.1186/s12984-020-00747-6
  4. Rama Bindiganavale and Norman I. Badler. 1998. Motion Abstraction and Mapping with Spatial Constraints. In Modelling and Motion Capture Techniques for Virtual Environments, Nadia Magnenat-Thalmann and Daniel Thalmann (Eds.). Springer Berlin Heidelberg, Berlin, Heidelberg, 70–82.
  5. Calculation of vertical ground reaction force estimates during running from positional data. Journal of Biomechanics 24, 12 (1991), 1095–1105. https://doi.org/10.1016/0021-9290(91)90002-5
  6. Estimating contact dynamics. In 2009 IEEE 12th International Conference on Computer Vision. 2389–2396. https://doi.org/10.1109/ICCV.2009.5459407
  7. OpenPose: realtime multi-person 2D pose estimation using Part Affinity Fields. In arXiv preprint arXiv:1812.08008.
  8. BodyPressure – Inferring Body Pose and Contact Pressure from a Depth Image. arXiv:2105.09936 [cs.CV]
  9. CMU CMU Graphics Lab. 2000. CMU Graphics Lab Motion Capture Database. http://mocap.cs.cmu.edu/
  10. Robert T. Collins. 2023. A Light-Weight Contrastive Approach for Aligning Human Pose Sequences. arXiv:2303.04244 [cs.CV]
  11. Ground reaction force patterns in knees with and without radiographic osteoarthritis and pain: descriptive analyses of a large cohort (the Multicenter Osteoarthritis Study). Osteoarthritis and Cartilage 29, 8 (2021), 1138–1146. https://doi.org/10.1016/j.joca.2021.03.009
  12. Relationship Between Jump-Landing Kinematics and Lower Extremity Overuse Injuries in Physically Active Populations: A Systematic Review and Meta-Analysis. Sports Medicine 50, 8 (2020), 1515–1532. https://doi.org/10.1007/s40279-020-01296-7 PMID: 32514700.
  13. Damiana A. dos Santos and Marcos Duarte. 2016. A Public Data set of Human Balance Evaluations. PeerJ Life and Environment (2016). https://doi.org/10.7717/peerj.2648
  14. A Data set with Kinematic and Ground Reaction Forces of Human Balance. PeerJ Life and Environment (2017). https://doi.org/10.7717/peerj.3626
  15. A public dataset of overground and treadmill walking kinematics and kinetics in healthy individuals. PeerJ Life and Environment (2018). https://doi.org/10.7717/peerj.4640
  16. A public dataset of running biomechanics and the effects of running speed on lower extremity kinematics and kinetics. PeerJ Life and Environment (2017). https://doi.org/10.7717/peerj.3298
  17. SuperTrack: Motion Tracking for Physically Simulated Characters Using Supervised Learning. ACM Trans. Graph. 40, 6, Article 197 (dec 2021), 13 pages. https://doi.org/10.1145/3478513.3480527
  18. PressureVision: Estimating Hand Pressure from a Single RGB Image. arXiv:2203.10385 [cs.CV]
  19. Samuel R. Hamner and Scott L. Delp. 2013. Muscle contributions to fore-aft and vertical body mass center accelerations over a range of running speeds. Journal of Biomechanics 46, 4 (2013), 780–787. https://doi.org/10.1016/j.jbiomech.2012.11.024
  20. Phase-Functioned Neural Networks for Character Control. ACM Trans. Graph. 36, 4, Article 42 (jul 2017), 13 pages. https://doi.org/10.1145/3072959.3073663
  21. A Deep Learning Framework for Character Motion Synthesis and Editing. ACM Trans. Graph. 35, 4, Article 138 (jul 2016), 11 pages. https://doi.org/10.1145/2897824.2925975
  22. Capturing and Inferring Dense Full-Body Human-Scene Contact. In Proceedings IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR). 13274–13285.
  23. A multimodal dataset of human gait at different walking speeds established on injury-free adult participants. Scientific Data 6, 1 (jul 2019), 111. https://doi.org/10.1038/s41597-019-0124-4
  24. Hyunho Jeong and Sukyung Park. 2020. Estimation of the ground reaction forces from a single video camera based on the spring-like center of mass dynamics of human walking. Journal of Biomechanics 113 (12 2020), 110074. https://doi.org/10.1016/j.jbiomech.2020.110074
  25. Predicting athlete ground reaction forces and moments from motion capture. Scientific Data 56, 10 (oct 2018), 1781–1792. https://doi.org/10.1007/s11517-018-1802-7
  26. Multidimensional Ground Reaction Forces and Moments From Wearable Sensor Accelerations via Deep Learning. IEEE Transactions on Biomedical Engineering 68, 1 (2021), 289–297. https://doi.org/10.1109/TBME.2020.3006158
  27. Estimation of Ground Reaction Forces and Moments During Gait Using Only Inertial Motion Capture. Sensors 17, 1 (2017). https://doi.org/10.3390/s17010075
  28. Estimation of the Knee Adduction Moment and Joint Contact Force during Daily Living Activities Using Inertial Motion Capture. Sensors 19 (04 2019), 1681. https://doi.org/10.3390/s19071681
  29. 4GAIT: Synchronized MoCap, Video, GRF and EMG Datasets: Acquisition, Management and Applications. 555–564. https://doi.org/10.1007/978-3-319-05458-2_57
  30. Fast and Flexible Multilegged Locomotion Using Learned Centroidal Dynamics. ACM Trans. Graph. 39, 4, Article 46 (aug 2020), 24 pages. https://doi.org/10.1145/3386569.3392432
  31. Interactive Control of Avatars Animated with Human Motion Data. ACM Trans. Graph. 21, 3 (jul 2002), 491–500. https://doi.org/10.1145/566654.566607
  32. Estimating 3D Motion and Forces of Person-Object Interactions from Monocular Video. In Computer Vision and Pattern Recognition (CVPR).
  33. Intelligent Carpet: Inferring 3D Human Pose from Tactile Signals. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 11250–11260.
  34. AMASS: Archive of Motion Capture as Surface Shapes. In International Conference on Computer Vision. 5442–5451.
  35. Ground reaction force metrics are not strongly correlated with tibial bone load when running across speeds and slopes: Implications for science, sport and wearable tech. PLoS ONE (2019). https://doi.org/10.1371/journal.pone.0210000
  36. An elaborate data set on human gait and the effect of mechanical perturbations. PeerJ Life and Environment (2015). https://doi.org/10.7717/peerj.918
  37. UnderPressure: Deep Learning for Foot Contact Detection, Ground Reaction Force Estimation and Footskate Cleanup. arXiv:2208.04598 [cs.GR]
  38. UnderPressure: Deep Learning of Foot Contacts and Forces for Character Animation. (2022). https://doi.org/10.48550/arXiv.2208.04598 arXiv:2208.04598 To be published in proceedings of Symposium on Computer Animation (SCA)..
  39. Estimating Ground Reaction Forces from Two-Dimensional Pose Data: A Biomechanics-Based Comparison of AlphaPose, BlazePose, and OpenPose. Sensors 23, 1 (2023). https://doi.org/10.3390/s23010078
  40. Prediction of ground reaction force and joint moments based on optical motion capture data during gait. Medical Engineering & Physics 86 (2020), 29–34. https://doi.org/10.1016/j.medengphy.2020.10.001
  41. The feasibility of predicting ground reaction forces during running from a trunk accelerometry driven mass-spring-damper model. PeerJ 6 (dec 2018), e6105. https://doi.org/10.7717/peerj.6105
  42. Expressive Body Capture: 3D Hands, Face, and Body from a Single Image. In Proceedings IEEE Conf. on Computer Vision and Pattern Recognition (CVPR). 10975–10985.
  43. Modeling Human Motion with Quaternion-Based Neural Networks. Int. J. Comput. Vision 128, 4 (apr 2020), 855–872. https://doi.org/10.1007/s11263-019-01245-6
  44. A neural network method to predict task- and step-specific ground reaction force magnitudes from trunk accelerations during running activities. Medical Engineering & Physics 78 (2020), 82–89. https://doi.org/10.1016/j.medengphy.2020.02.002
  45. Qualisys. 2021. Qualisys Motion Capture System. https://www.qualisys.com/ Accessed: 2022-05-30.
  46. Center of pressure displacement characteristics differentiate fall risk in older people: A systematic review with meta-analysis. Ageing Research Reviews 62 (2020), 101117. https://doi.org/10.1016/j.arr.2020.101117 PMID: 32565327.
  47. Estimation of ground reaction forces from kinematics in ski-jumping imitation jumps.
  48. From Image to Stability: Learning Dynamics from Human Pose. In Computer Vision – ECCV 2020, Andrea Vedaldi, Horst Bischof, Thomas Brox, and Jan-Michael Frahm (Eds.). Springer International Publishing, Cham, 536–554. https://doi.org/10.1007/978-3-030-58592-1_32
  49. Erfan Shahabpoor and Aleksandar Pavic. 2017. Measurement of Walking Ground Reactions in Real-Life Environments: A Systematic Review of Techniques and Technologies. Sensors (Basel) 17, 9 (Sep 2017), 2085. https://doi.org/10.3390/s17092085 PMCID: PMC5620730.
  50. Indirect Estimation of Vertical Ground Reaction Force from a Body-Mounted INS/GPS Using Machine Learning. Sensors 21 (02 2021), 1553. https://doi.org/10.3390/s21041553
  51. HULC: 3D HUman Motion Capture with Pose Manifold SampLing and Dense Contact Guidance. In European Conference on Computer Vision (ECCV). 516–533.
  52. Neural Monocular 3D Human Motion Capture with Physical Awareness. ACM Transactions on Graphics 40, 4, Article 83 (aug 2021).
  53. PhysCap: Physically Plausible Monocular 3D Motion Capture in Real Time. ACM Trans. Graph. 39, 6, Article 235 (nov 2020), 16 pages. https://doi.org/10.1145/3414685.3417877
  54. Prediction of ground reaction forces and moments during sports-related movements. Multibody System Dynamics 39 (03 2017). https://doi.org/10.1007/s11044-016-9537-4
  55. Neural State Machine for Character-Scene Interactions. ACM Trans. Graph. 38, 6, Article 209 (nov 2019), 14 pages. https://doi.org/10.1145/3355089.3356505
  56. Total Capture: 3D Human Pose Estimation Fusing Video and Inertial Sensors. In 2017 British Machine Vision Conference (BMVC).
  57. Do runners who suffer injuries have higher vertical ground reaction forces than those who remain injury-free? A systematic review and meta-analysis. British Journal of Sports Medicine 50, 8 (2016), 450–457. https://doi.org/10.1136/bjsports-2015-094924 PMID: 26729857.
  58. Whole-body biomechanical load in running-based sports: The validity of estimating ground reaction forces from segmental accelerations. Journal of Science and Medicine in Sport 22, 6 (2019), 716–722. https://doi.org/10.1016/j.jsams.2018.12.007
  59. Huawei Wang and Antonie J. van den Bogert. 2020. Identification of the human postural control system through stochastic trajectory optimization. Journal of Neuroscience Methods 334 (2020), 108580. https://doi.org/10.1016/j.jneumeth.2020.108580
  60. Huawei Wang and Antonie J. van den Bogert. 2020. Standing Balance Experiment with Long Duration Random Pulses Perturbation. https://api.semanticscholar.org/CorpusID:244988575
  61. Jungdam Won and J. Lee. 2019. Learning body shape variation in physics-based characters. ACM Transactions on Graphics 38 (11 2019), 1–12. https://doi.org/10.1145/3355089.3356499
  62. Estimation of Vertical Ground Reaction Forces and Sagittal Knee Kinematics During Running Using Three Inertial Sensors. Frontiers in Physiology 9 (03 2018). https://doi.org/10.3389/fphys.2018.00218
  63. KangKang Yin and Dinesh K. Pai. 2003. FootSee: an interactive animation system. In Symposium on Computer Animation. https://api.semanticscholar.org/CorpusID:14380898
  64. KangKang Yin and Goh Jing Ying. 2021. SFU Motion Capture Database. https://mocap.cs.sfu.ca/ Accessed: 2023-05-01.
  65. Joint 3D Human Motion Capture and Physical Analysis from Monocular Videos. 17–26. https://doi.org/10.1109/CVPRW.2017.9
  66. Mode-Adaptive Neural Networks for Quadruped Motion Control. ACM Trans. Graph. 37, 4, Article 145 (jul 2018), 11 pages. https://doi.org/10.1145/3197517.3201366
Citations (6)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets