pith. sign in

arxiv: 1907.10226 · v1 · pith:DA5KGM5Vnew · submitted 2019-07-24 · 💻 cs.CV · q-bio.QM

Movement science needs different pose tracking algorithms

Nidhi Seethapathi , Shaofei Wang , Rachit Saluja , Gunnar Blohm , Konrad P. Kording This is my paper

Pith reviewed 2026-05-24 17:16 UTC · model grok-4.3

classification 💻 cs.CV q-bio.QM
keywords pose trackingmovement sciencekinematics3D pose estimationevaluation metricsbiomechanicscomputer visionforce estimation
0
0 comments X

The pith

Current pose tracking algorithms do not meet the precision needs of movement science for kinematic data.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper argues that progress in extracting body pose from video has not translated to movement science because evaluation metrics rely on noisy hand-labeled data and overlook the accuracy required for three-dimensional position, velocity, acceleration, and forces. Movement science disciplines use these variables to detect disease, quantify performance, and study motion outside labs. The authors introduce the data needs across relevant fields and outline the adjustments required in tracking development and testing to make the technology useful.

Core claim

Pose tracking algorithms fall short of movement science needs because they poorly estimate the types of movement data that matter, such as three-dimensional position, velocity, acceleration, and forces. Evaluation metrics rely on noisy hand-labeled ground truth and do not prioritize the precision of these variables, which are crucial for detecting disease, quantifying performance, and moving the science out of the lab.

What carries the argument

Evaluation metrics based on noisy hand-labeled data that fail to prioritize kinematic precision and force estimation.

If this is right

  • Pose tracking could enable disease detection from everyday videos if metrics align with needed variables.
  • Quantification of movement performance would improve with accurate acceleration data.
  • Studies could move out of labs into real-world settings with better force and velocity estimates.
  • New algorithms would need to be designed with movement science requirements in mind.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Improving metrics might indirectly address reconstruction issues by guiding algorithm development toward 3D accuracy.
  • Integration with other sensors could become a focus if kinematics alone prove insufficient.
  • This mismatch may explain why pose tracking has seen limited adoption in biomechanics despite computer vision advances.

Load-bearing premise

That changing evaluation metrics is the main change needed to make pose tracking useful for movement science, as opposed to fundamental limits in recovering 3D data or forces from video.

What would settle it

An experiment showing that even with metrics prioritizing velocity and acceleration, pose tracking still cannot provide reliable data for a specific movement science application due to video limitations.

Figures

Figures reproduced from arXiv: 1907.10226 by Gunnar Blohm, Konrad P. Kording, Nidhi Seethapathi, Rachit Saluja, Shaofei Wang.

Figure 1
Figure 1. Figure 1: Many important disciplines of science and engineering rely on human movement data for research. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The need for better pose tracking algorithms. a) Most pose estimation papers published in computer vision conferences in the past two years do not use temporal information. b) Typical failure modes when algorithms are applied to videos of interest to movement science. In these cases, the algorithm’s performance is clearly inferior to that of the human eye. To generate this figure, we processed a video of a… view at source ↗
Figure 3
Figure 3. Figure 3: Key takeaways from the paper regarding what movement science needs from pose tracking and how to get there. algorithms would not be good for. Similarly, in-the-wild data consists of occlusions which currently cause some pose tracking algorithms to fail [87] with the only solution being to handpick video frames where occlusions are absent. These issues with not detecting contact and partial occlusions can b… view at source ↗
read the original abstract

Over the last decade, computer science has made progress towards extracting body pose from single camera photographs or videos. This promises to enable movement science to detect disease, quantify movement performance, and take the science out of the lab into the real world. However, current pose tracking algorithms fall short of the needs of movement science; the types of movement data that matter are poorly estimated. For instance, the metrics currently used for evaluating pose tracking algorithms use noisy hand-labeled ground truth data and do not prioritize precision of relevant variables like three-dimensional position, velocity, acceleration, and forces which are crucial for movement science. Here, we introduce the scientific disciplines that use movement data, the types of data they need, and discuss the changes needed to make pose tracking truly transformative for movement science.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper argues that computer-vision pose tracking algorithms fall short for movement science because their evaluation metrics rely on noisy hand-labeled ground truth and fail to prioritize accuracy on the 3D position, velocity, acceleration, and force variables that movement scientists require. It introduces the relevant scientific disciplines, describes the data types they need, and outlines changes to pose tracking that would make the technology transformative outside the lab.

Significance. If the argument is substantiated, the manuscript could usefully redirect algorithm development toward biomechanically relevant error measures and thereby improve the applicability of monocular pose tracking to clinical and performance analysis. The absence of any quantitative comparison of current estimators on velocity/acceleration/force error versus hand-labeled references, however, leaves the central claim as an assertion rather than a demonstrated gap.

major comments (2)
  1. [Abstract] Abstract, paragraph 3: the claim that redesigning evaluation metrics will close the gap between pose tracking and movement-science needs assumes that current estimators already encode the required kinematic and dynamic information; the manuscript provides no evidence that this is the case and does not engage with the scale/depth ambiguity inherent in monocular reconstruction or the under-constrained nature of force recovery from kinematics alone.
  2. [Abstract] Abstract: no quantitative comparison is supplied between existing pose-tracking error on velocity, acceleration, or force estimates and hand-labeled ground truth, so the assertion that 'the types of movement data that matter are poorly estimated' rests on qualitative statement rather than measured discrepancy.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the detailed and constructive report. Our manuscript is a perspective piece whose goal is to describe the data requirements of movement science and to argue for changes in how pose-tracking algorithms are evaluated and developed. We respond to each major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract, paragraph 3: the claim that redesigning evaluation metrics will close the gap between pose tracking and movement-science needs assumes that current estimators already encode the required kinematic and dynamic information; the manuscript provides no evidence that this is the case and does not engage with the scale/depth ambiguity inherent in monocular reconstruction or the under-constrained nature of force recovery from kinematics alone.

    Authors: The manuscript does not claim that redesigning metrics alone will close the gap or that current estimators already encode the needed information. Its argument is that evaluation criteria should be aligned with the kinematic and kinetic variables required by movement scientists; such alignment would create incentives for algorithmic improvements that address monocular ambiguities and force estimation where feasible. We agree the text does not analyze scale/depth ambiguity or the under-constrained inverse-dynamics problem in detail. We will revise the abstract and relevant sections to state explicitly that metric redesign is a necessary but not necessarily sufficient step. revision: partial

  2. Referee: [Abstract] Abstract: no quantitative comparison is supplied between existing pose-tracking error on velocity, acceleration, or force estimates and hand-labeled ground truth, so the assertion that 'the types of movement data that matter are poorly estimated' rests on qualitative statement rather than measured discrepancy.

    Authors: The manuscript supplies no quantitative comparison because it is a perspective article, not an empirical benchmarking study. The central claim rests on the documented mismatch between (a) standard pose-tracking metrics that rely on noisy 2-D/3-D keypoint labels and (b) the precision movement scientists require for derived quantities such as velocity, acceleration, and forces. We maintain that this qualitative diagnosis is appropriate for the paper’s scope; performing the requested quantitative analysis would constitute a separate experimental contribution. revision: no

standing simulated objections not resolved
  • Absence of any quantitative comparison of current estimators on velocity/acceleration/force error versus hand-labeled references.

Circularity Check

0 steps flagged

No circularity: direct domain-mismatch argument with no derivations or self-referential reductions

full rationale

The paper is a perspective piece arguing that pose-tracking evaluation metrics are mismatched to movement-science needs (noisy labels, failure to prioritize 3D kinematics/forces). It contains no equations, fitted parameters, predictions, or derivation chain. The central claim is a direct statement of requirements rather than a result obtained by construction from its own inputs. No self-citation load-bearing steps, ansatzes, or renamings are present. This matches the default expectation of non-circularity for non-technical arguments.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No mathematical model, derivation, or empirical result is presented in the abstract; the argument rests on domain knowledge about what movement scientists need.

pith-pipeline@v0.9.0 · 5669 in / 1094 out tokens · 16117 ms · 2026-05-24T17:16:00.663949+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

92 extracted references · 92 canonical work pages · 5 internal anchors

  1. [1]

    Neural population dynamics during reaching

    Mark M Churchland, John P Cunningham, Matthew T Kaufman, Justin D Foster, Paul Nuyujukian, Stephen I Ryu, and Krishna V Shenoy. Neural population dynamics during reaching. Nature, 487(7405):51, 2012

    work page 2012

  2. [2]

    Target selection signals for arm reaching in the posterior parietal cortex

    Hansjörg Scherberger and Richard A Andersen. Target selection signals for arm reaching in the posterior parietal cortex. Journal of Neuroscience, 27(8):2001–2012, 2007

    work page 2001

  3. [3]

    Neural correlates of cognitive control of reaching movements in the dorsal premotor cortex of rhesus monkeys

    Giovanni Mirabella, Pierpaolo Pani, and Stefano Ferraina. Neural correlates of cognitive control of reaching movements in the dorsal premotor cortex of rhesus monkeys. American Journal of Physiology-Heart and Circulatory Physiology, 2011

    work page 2011

  4. [4]

    Computer optimization of a minimal biped model discovers walking and running

    Manoj Srinivasan and Andy Ruina. Computer optimization of a minimal biped model discovers walking and running. Nature, 439(7072):72, 2006

    work page 2006

  5. [5]

    Opensim: open-source software to create and analyze dynamic simulations of movement

    Scott L Delp, Frank C Anderson, Allison S Arnold, Peter Loan, Ayman Habib, Chand T John, Eran Guendelman, and Darryl G Thelen. Opensim: open-source software to create and analyze dynamic simulations of movement. IEEE transactions on biomedical engineering, 54(11):1940–1950, 2007

    work page 1940

  6. [6]

    Walking patterns and hip contact forces in patients with hip dysplasia

    Ole Skalshøi, Christian Hauskov Iversen, Dennis Brandborg Nielsen, Julie Jacobsen, Inger Mechlenburg, Kjeld Søballe, and Henrik Sørensen. Walking patterns and hip contact forces in patients with hip dysplasia. Gait & posture, 42(4):529–533, 2015

    work page 2015

  7. [7]

    Step-to-step variations in human running reveal how humans run without falling

    Nidhi Seethapathi and Manoj Srinivasan. Step-to-step variations in human running reveal how humans run without falling. eLife, 8:e38371, 2019

    work page 2019

  8. [8]

    Rectus femoris transfer surgery affects balance recovery in children with cerebral palsy: a computer simulation study

    Misagh Mansouri, Ashley E Clark, Ajay Seth, and Jeffrey A Reinbolt. Rectus femoris transfer surgery affects balance recovery in children with cerebral palsy: a computer simulation study. Gait & posture, 43:24–30, 2016

    work page 2016

  9. [9]

    Ambulatory assessment of ankle and foot dynamics

    H Martin Schepers, Hubertus FJM Koopman, and Peter H Veltink. Ambulatory assessment of ankle and foot dynamics. IEEE Transactions on Biomedical Engineering, 54(5):895–902, 2007

    work page 2007

  10. [10]

    Gauging force by tapping tendons

    Jack A Martin, Scott CE Brandon, Emily M Keuler, James R Hermus, Alexander C Ehlers, Daniel J Segalman, Matthew S Allen, and Darryl G Thelen. Gauging force by tapping tendons. Nature communications, 9(1):1592, 2018

    work page 2018

  11. [11]

    The effects of resisted sprint training on acceleration performance and kinematics in soccer, rugby union, and australian football players

    Christopher D Spinks, Aron J Murphy, Warwick L Spinks, and Robert G Lockie. The effects of resisted sprint training on acceleration performance and kinematics in soccer, rugby union, and australian football players. The Journal of Strength & Conditioning Research, 21(1):77–85, 2007

    work page 2007

  12. [12]

    Acceleration profiles in elite australian soccer.International journal of sports medicine, 34(01):34–39, 2013

    Matthew C Varley and Robert J Aughey. Acceleration profiles in elite australian soccer.International journal of sports medicine, 34(01):34–39, 2013

    work page 2013

  13. [13]

    Age-related differences in acceleration, maximum running speed, and repeated-sprint performance in young soccer players

    Alberto Mendez-Villanueva, Martin Buchheit, Sami Kuitunen, Andrew Douglas, Esa Peltola, and Pitre Bourdon. Age-related differences in acceleration, maximum running speed, and repeated-sprint performance in young soccer players. Journal of sports sciences, 29(5):477–484, 2011

    work page 2011

  14. [14]

    Evaluating sprinting ability, density of acceleration, and speed dribbling ability of professional soccer players with respect to their positions

    Halil Taskin. Evaluating sprinting ability, density of acceleration, and speed dribbling ability of professional soccer players with respect to their positions. The Journal of Strength & Conditioning Research, 22(5):1481–1486, 2008

    work page 2008

  15. [15]

    Walking on music

    Frederik Styns, Leon van Noorden, Dirk Moelants, and Marc Leman. Walking on music. Human movement science, 26(5):769–785, 2007. 9

    work page 2007

  16. [16]

    Follow the leader: Visual control of speed in pedestrian following

    Kevin W Rio, Christopher K Rhea, and William H Warren. Follow the leader: Visual control of speed in pedestrian following. Journal of vision, 14(2):4–4, 2014

    work page 2014

  17. [17]

    The direction of walking—but not throwing or kicking—is adapted by optic flow

    Hugo Bruggeman and William H Warren. The direction of walking—but not throwing or kicking—is adapted by optic flow. Psychological Science, 21(7):1006–1013, 2010

    work page 2010

  18. [18]

    Fear of heights in infants?Current directions in psychological science, 23(1):60–66, 2014

    Karen E Adolph, Kari S Kretch, and Vanessa LoBue. Fear of heights in infants?Current directions in psychological science, 23(1):60–66, 2014

    work page 2014

  19. [19]

    How do you learn to walk? thousands of steps and dozens of falls per day

    Karen E Adolph, Whitney G Cole, Meghana Komati, Jessie S Garciaguirre, Daryaneh Badaly, Jesse M Lingeman, Gladys LY Chan, and Rachel B Sotsky. How do you learn to walk? thousands of steps and dozens of falls per day. Psychological science, 23(11):1387–1394, 2012

    work page 2012

  20. [20]

    Postural activation of the human medial gastrocnemius muscle: are the muscle units spatially localised? The Journal of physiology , 589(2):431–443, 2011

    Taian MM Vieira, Ian D Loram, Silvia Muceli, Roberto Merletti, and Dario Farina. Postural activation of the human medial gastrocnemius muscle: are the muscle units spatially localised? The Journal of physiology , 589(2):431–443, 2011

    work page 2011

  21. [21]

    The passive, human calf muscles in relation to standing: the non-linear decrease from short range to long range stiffness

    Ian D Loram, Constantinos N Maganaris, and Martin Lakie. The passive, human calf muscles in relation to standing: the non-linear decrease from short range to long range stiffness. The Journal of physiology, 584(2):661– 675, 2007

    work page 2007

  22. [22]

    Mechanical and neural stretch responses of the human soleus muscle at different walking speeds

    Neil J Cronin, Masaki Ishikawa, Michael J Grey, Richard Af Klint, Paavo V Komi, Janne Avela, Thomas Sinkjaer, and Michael V oigt. Mechanical and neural stretch responses of the human soleus muscle at different walking speeds. The Journal of physiology, 587(13):3375–3382, 2009

    work page 2009

  23. [23]

    Muscle synergies characterizing human postural responses

    Gelsy Torres-Oviedo and Lena H Ting. Muscle synergies characterizing human postural responses. Journal of neurophysiology, 2007

    work page 2007

  24. [24]

    Control of position and movement is simplified by combined muscle spindle and golgi tendon organ feedback.American Journal of Physiology-Heart and Circulatory Physiology, 2012

    Dinant Arne Kistemaker, Arthur Knoek J Van Soest, Jeremy D Wong, Isaac L Kurtzer, and Paul L Gribble. Control of position and movement is simplified by combined muscle spindle and golgi tendon organ feedback.American Journal of Physiology-Heart and Circulatory Physiology, 2012

    work page 2012

  25. [25]

    Locomotion speed determines gait variability in cerebellar ataxia and vestibular failure

    Roman Schniepp, Maximilian Wuehr, Maximilian Neuhaeusser, Maria Kamenova, Konstantin Dimitriadis, Thomas Klopstock, M Strupp, Thomas Brandt, and Klaus Jahn. Locomotion speed determines gait variability in cerebellar ataxia and vestibular failure. Movement disorders, 27(1):125–131, 2012

    work page 2012

  26. [26]

    Stepping in the direction of the fall: the next foot placement can be predicted from current upper body state in steady-state walking

    Yang Wang and Manoj Srinivasan. Stepping in the direction of the fall: the next foot placement can be predicted from current upper body state in steady-state walking. Biology letters, 10(9):20140405, 2014

    work page 2014

  27. [27]

    Levodopa influences the regularity of the ankle joint kinematics in individuals with parkinson’s disease

    Max J Kurz and Jyhgong Gabriel Hou. Levodopa influences the regularity of the ankle joint kinematics in individuals with parkinson’s disease. Journal of computational neuroscience, 28(1):131–136, 2010

    work page 2010

  28. [28]

    Effects of bilateral pallidal or subthalamic stimulation on gait in advanced parkinson’s disease

    N Allert, J V olkmann, S Dotse, H Hefter, V Sturm, and H-J Freund. Effects of bilateral pallidal or subthalamic stimulation on gait in advanced parkinson’s disease. Movement disorders: official journal of the Movement Disorder Society, 16(6):1076–1085, 2001

    work page 2001

  29. [29]

    Online reinforcement learning control for the personalization of a robotic knee prosthesis

    Yue Wen, Jennie Si, Andrea Brandt, Xiang Gao, and He Huang. Online reinforcement learning control for the personalization of a robotic knee prosthesis. IEEE transactions on cybernetics, 2019

    work page 2019

  30. [30]

    Human-in-the-loop optimization of exoskeleton assistance during walking

    Juanjuan Zhang, Pieter Fiers, Kirby A Witte, Rachel W Jackson, Katherine L Poggensee, Christopher G Atkeson, and Steven H Collins. Human-in-the-loop optimization of exoskeleton assistance during walking. Science, 356(6344):1280–1284, 2017

    work page 2017

  31. [31]

    A minimum energy cost hypothesis for human arm trajectories

    R McN Alexander. A minimum energy cost hypothesis for human arm trajectories. Biological cybernetics, 76(2):97–105, 1997

    work page 1997

  32. [32]

    Evidence for the flexible sensorimotor strategies predicted by optimal feedback control

    Dan Liu and Emanuel Todorov. Evidence for the flexible sensorimotor strategies predicted by optimal feedback control. Journal of Neuroscience, 27(35):9354–9368, 2007

    work page 2007

  33. [33]

    Causal inference in multisensory perception

    Konrad P Körding, Ulrik Beierholm, Wei Ji Ma, Steven Quartz, Joshua B Tenenbaum, and Ladan Shams. Causal inference in multisensory perception. PLoS one, 2(9):e943, 2007

    work page 2007

  34. [34]

    Optimality principles for model-based prediction of human gait

    Marko Ackermann and Antonie J Van den Bogert. Optimality principles for model-based prediction of human gait. Journal of biomechanics, 43(6):1055–1060, 2010

    work page 2010

  35. [35]

    The metabolic cost of changing walking speeds is significant, implies lower optimal speeds for shorter distances, and increases daily energy estimates

    Nidhi Seethapathi and Manoj Srinivasan. The metabolic cost of changing walking speeds is significant, implies lower optimal speeds for shorter distances, and increases daily energy estimates. Biology letters, 11(9):20150486, 2015

    work page 2015

  36. [36]

    Mechanical and metabolic determinants of the preferred step width in human walking

    J Maxwell Donelan, Rodger Kram, and Kuo Arthur D. Mechanical and metabolic determinants of the preferred step width in human walking. Proceedings of the Royal Society of London. Series B: Biological Sciences , 268(1480):1985–1992, 2001. 10

    work page 1985

  37. [37]

    Multiple walking speed–frequency relations are predicted by constrained optimization

    John EA Bertram and Andy Ruina. Multiple walking speed–frequency relations are predicted by constrained optimization. Journal of theoretical Biology, 209(4):445–453, 2001

    work page 2001

  38. [38]

    Is slow walking more stable? Journal of biomechanics, 42(10):1506–1512, 2009

    Sjoerd M Bruijn, Jaap H van Dieën, Onno G Meijer, and Peter J Beek. Is slow walking more stable? Journal of biomechanics, 42(10):1506–1512, 2009

    work page 2009

  39. [39]

    The effect of lateral stabilization on walking in young and old adults

    Jesse C Dean, Neil B Alexander, and Arthur D Kuo. The effect of lateral stabilization on walking in young and old adults. IEEE Transactions on Biomedical Engineering, 54(11):1919–1926, 2007

    work page 1919

  40. [40]

    Bio-inspired swing leg control for spring-mass robots running on ground with unexpected height disturbance

    HR Vejdani, Y Blum, MA Daley, and JW Hurst. Bio-inspired swing leg control for spring-mass robots running on ground with unexpected height disturbance. Bioinspiration & biomimetics, 8(4):046006, 2013

    work page 2013

  41. [41]

    Locomotion among dynamic obstacles for the honda asimo

    Joel Chestnutt, Philipp Michel, James Kuffner, and Takeo Kanade. Locomotion among dynamic obstacles for the honda asimo. In 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, pages 2572–2573. IEEE, 2007

    work page 2007

  42. [42]

    Design and control of ranger: an energy-efficient, dynamic walking robot

    Pranav A Bhounsule, Jason Cortell, and Andy Ruina. Design and control of ranger: an energy-efficient, dynamic walking robot. In Adaptive Mobile Robotics, pages 441–448. World Scientific, 2012

    work page 2012

  43. [43]

    Multi-joint actuation platform for lower extremity soft exosuits

    Ye Ding, Ignacio Galiana, Alan Asbeck, Brendan Quinlivan, Stefano Marco Maria De Rossi, and Conor Walsh. Multi-joint actuation platform for lower extremity soft exosuits. In 2014 IEEE International Conference on Robotics and Automation (ICRA), pages 1327–1334. Ieee, 2014

    work page 2014

  44. [44]

    Potential of a suite of robot/computer- assisted motivating systems for personalized, home-based, stroke rehabilitation

    Michelle J Johnson, Xin Feng, Laura M Johnson, and Jack M Winters. Potential of a suite of robot/computer- assisted motivating systems for personalized, home-based, stroke rehabilitation. Journal of NeuroEngineering and Rehabilitation, 4(1):6, 2007

    work page 2007

  45. [45]

    Expression of emotion in the kinematics of locomotion

    Avi Barliya, Lars Omlor, Martin A Giese, Alain Berthoz, and Tamar Flash. Expression of emotion in the kinematics of locomotion. Experimental brain research, 225(2):159–176, 2013

    work page 2013

  46. [46]

    Perception of emotional gaits using avatar animation of real and artificially synthesized gaits

    Halim Hicheur, Hideki Kadone, Julie Grezes, and Alain Berthoz. Perception of emotional gaits using avatar animation of real and artificially synthesized gaits. In 2013 Humaine Association Conference on Affective Computing and Intelligent Interaction, pages 460–466. IEEE, 2013

    work page 2013

  47. [47]

    Communicating emotions and mental states to robots in a real time parallel framework using laban movement analysis

    Tino Lourens, Roos Van Berkel, and Emilia Barakova. Communicating emotions and mental states to robots in a real time parallel framework using laban movement analysis. Robotics and Autonomous Systems, 58(12):1256– 1265, 2010

    work page 2010

  48. [48]

    Scaling of sensorimotor control in terrestrial mammals

    Heather L More, John R Hutchinson, David F Collins, Douglas J Weber, Steven KH Aung, and J Maxwell Donelan. Scaling of sensorimotor control in terrestrial mammals. Proceedings of the Royal Society B: Biological Sciences, 277(1700):3563–3568, 2010

    work page 2010

  49. [49]

    Compass gait mechanics account for top walking speeds in ducks and humans

    James R Usherwood, Katie L Szymanek, and Monica A Daley. Compass gait mechanics account for top walking speeds in ducks and humans. Journal of Experimental Biology, 211(23):3744–3749, 2008

    work page 2008

  50. [50]

    Human recognition based on kinematics and kinetics of gait

    Y Chih Lin, B Shiang Yang, Yu Tzu Lin, Yi Ting Yang, et al. Human recognition based on kinematics and kinetics of gait. Journal of Medical and Biological Engineering, 31(4):255–263, 2011

    work page 2011

  51. [51]

    Learning human identity from motion patterns

    Natalia Neverova, Christian Wolf, Griffin Lacey, Lex Fridman, Deepak Chandra, Brandon Barbello, and Graham Taylor. Learning human identity from motion patterns. IEEE Access, 4:1810–1820, 2016

    work page 2016

  52. [52]

    Behavioral tracking gets real

    Kunlin Wei and Konrad Paul Kording. Behavioral tracking gets real. Nature neuroscience, 21(9):1146, 2018

    work page 2018

  53. [53]

    Pose estimates from online videos show that side-by-side walkers synchronize movement under naturalistic conditions

    Claire Chambers, Gaiqing Kong, Kunlin Wei, and Konrad Kording. Pose estimates from online videos show that side-by-side walkers synchronize movement under naturalistic conditions. PloS one, 14(6):e0217861, 2019

    work page 2019

  54. [54]

    Catalin Ionescu, Dragos Papava, Vlad Olaru, and Cristian Sminchisescu. Human3. 6m: Large scale datasets and predictive methods for 3d human sensing in natural environments. IEEE transactions on pattern analysis and machine intelligence, 36(7):1325–1339, 2013

    work page 2013

  55. [55]

    Microsoft coco: Common objects in context

    Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár, and C Lawrence Zitnick. Microsoft coco: Common objects in context. In European conference on computer vision, pages 740–755. Springer, 2014

    work page 2014

  56. [56]

    Posetrack: A benchmark for human pose estimation and tracking

    Mykhaylo Andriluka, Umar Iqbal, Eldar Insafutdinov, Leonid Pishchulin, Anton Milan, Juergen Gall, and Bernt Schiele. Posetrack: A benchmark for human pose estimation and tracking. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 5167–5176, 2018

    work page 2018

  57. [57]

    A kinematic and electromyographic analysis of turning in people with parkinson disease

    Minna Hong, Joel S Perlmutter, and Gammon M Earhart. A kinematic and electromyographic analysis of turning in people with parkinson disease. Neurorehabilitation and neural repair, 23(2):166–176, 2009

    work page 2009

  58. [58]

    Deep High-Resolution Representation Learning for Human Pose Estimation

    Ke Sun, Bin Xiao, Dong Liu, and Jingdong Wang. Deep high-resolution representation learning for human pose estimation. arXiv preprint arXiv:1902.09212, 2019. 11

    work page internal anchor Pith review Pith/arXiv arXiv 1902

  59. [59]

    Learning feature pyramids for human pose estimation

    Wei Yang, Shuang Li, Wanli Ouyang, Hongsheng Li, and Xiaogang Wang. Learning feature pyramids for human pose estimation. In Proceedings of the IEEE International Conference on Computer Vision, pages 1281–1290, 2017

    work page 2017

  60. [60]

    Stacked hourglass networks for human pose estimation

    Alejandro Newell, Kaiyu Yang, and Jia Deng. Stacked hourglass networks for human pose estimation. InEuropean Conference on Computer Vision, pages 483–499. Springer, 2016

    work page 2016

  61. [61]

    Parsing R-CNN for Instance-Level Human Analysis

    Lu Yang, Qing Song, Zhihui Wang, and Ming Jiang. Parsing r-cnn for instance-level human analysis. arXiv preprint arXiv:1811.12596, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  62. [62]

    Self-Supervised Learning of 3D Human Pose using Multi-view Geometry

    Muhammed Kocabas, Salih Karagoz, and Emre Akbas. Self-supervised learning of 3d human pose using multi-view geometry. arXiv preprint arXiv:1903.02330, 2019

    work page internal anchor Pith review Pith/arXiv arXiv 1903

  63. [63]

    An Integral Pose Regression System for the ECCV2018 PoseTrack Challenge

    Xiao Sun, Chuankang Li, and Stephen Lin. An integral pose regression system for the eccv2018 posetrack challenge. arXiv preprint arXiv:1809.06079, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  64. [64]

    Unsupervised learning of depth and ego- motion from video

    Tinghui Zhou, Matthew Brown, Noah Snavely, and David G Lowe. Unsupervised learning of depth and ego- motion from video. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 1851–1858, 2017

    work page 2017

  65. [65]

    SfM-Net: Learning of Structure and Motion from Video

    Sudheendra Vijayanarasimhan, Susanna Ricco, Cordelia Schmid, Rahul Sukthankar, and Katerina Fragkiadaki. Sfm-net: Learning of structure and motion from video. arXiv preprint arXiv:1704.07804, 2017

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  66. [66]

    Densepose: Dense human pose estimation in the wild

    Rıza Alp Güler, Natalia Neverova, and Iasonas Kokkinos. Densepose: Dense human pose estimation in the wild. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 7297–7306, 2018

    work page 2018

  67. [67]

    Data collection in a flat world: The strengths and weaknesses of mechanical turk samples

    Joseph K Goodman, Cynthia E Cryder, and Amar Cheema. Data collection in a flat world: The strengths and weaknesses of mechanical turk samples. Journal of Behavioral Decision Making, 26(3):213–224, 2013

    work page 2013

  68. [68]

    Catalin Ionescu, Dragos Papava, Vlad Olaru, and Cristian Sminchisescu. Human3. 6m: Large scale datasets and predictive methods for 3d human sensing in natural environments. IEEE transactions on pattern analysis and machine intelligence, 36(7):1325–1339, 2014

    work page 2014

  69. [69]

    Detectron

    Ross Girshick, Ilija Radosavovic, Georgia Gkioxari, Piotr Dollár, and Kaiming He. Detectron. https://github. com/facebookresearch/detectron, 2018

    work page 2018

  70. [70]

    Biomechanical response to hamstring muscle strain injury

    Anthony G Schache, Tim V Wrigley, Richard Baker, and Marcus G Pandy. Biomechanical response to hamstring muscle strain injury. Gait & posture, 29(2):332–338, 2009

    work page 2009

  71. [71]

    Graphical interpretation of the solution to the redundant problem in biomechanics

    EY Chao and KN An. Graphical interpretation of the solution to the redundant problem in biomechanics. Journal of Biomechanical Engineering, 100(3):159–167, 1978

    work page 1978

  72. [72]

    Frequency domain characteristics of ground reaction forces during walking of young and elderly females

    Nicholas Stergiou, Giannis Giakas, Jennifer E Byrne, and Valerie Pomeroy. Frequency domain characteristics of ground reaction forces during walking of young and elderly females. Clinical Biomechanics, 17(8):615–617, 2002

    work page 2002

  73. [73]

    Martin H. Weik. Nyquist theorem, pages 1127–1127. Springer US, Boston, MA, 2001

    work page 2001

  74. [74]

    Ball–foot interaction in impact phase of instep soccer kicking

    Hironari Shinkai, Hiroyuki Nunome, Yasuo Ikegami, and Masanori Isokawa. Ball–foot interaction in impact phase of instep soccer kicking. Science and football VI, 6:41, 2008

    work page 2008

  75. [75]

    How children move: activity pattern characteris- tics in lean and obese chinese children

    Alison M McManus, Eva YW Chu, Clare CW Yu, and Yong Hu. How children move: activity pattern characteris- tics in lean and obese chinese children. Journal of obesity, 2011, 2010

    work page 2011

  76. [76]

    Whole body inverse dynamics over a complete gait cycle based only on measured kinematics

    Lei Ren, Richard K Jones, and David Howard. Whole body inverse dynamics over a complete gait cycle based only on measured kinematics. Journal of biomechanics, 41(12):2750–2759, 2008

    work page 2008

  77. [77]

    Adjustments to mcconville et al

    Raphael Dumas, Laurence Cheze, and J-P Verriest. Adjustments to mcconville et al. and young et al. body segment inertial parameters. Journal of biomechanics, 40(3):543–553, 2007

    work page 2007

  78. [78]

    Age-related differences in spatiotem- poral markers of gait stability during dual task walking

    John H Hollman, Francine M Kovash, Jared J Kubik, and Rachel A Linbo. Age-related differences in spatiotem- poral markers of gait stability during dual task walking. Gait & posture, 26(1):113–119, 2007

    work page 2007

  79. [79]

    Falls in the elderly: what can be done? International nursing review, 54(2):191–196, 2007

    AD Akyol. Falls in the elderly: what can be done? International nursing review, 54(2):191–196, 2007

    work page 2007

  80. [80]

    Comparison of landing biomechanics between male and female dancers and athletes, part 2: influence of fatigue and implications for anterior cruciate ligament injury

    Marijeanne Liederbach, Ian J Kremenic, Karl F Orishimo, Evangelos Pappas, and Marshall Hagins. Comparison of landing biomechanics between male and female dancers and athletes, part 2: influence of fatigue and implications for anterior cruciate ligament injury. The American journal of sports medicine, 42(5):1089–1095, 2014

    work page 2014

Showing first 80 references.