pith. sign in

arxiv: 2502.19056 · v2 · submitted 2025-02-26 · 💻 cs.GR · cs.LG

Fatigue-PINN: Physics-Informed Fatigue-Driven Motion Modulation and Synthesis

Iliana Loi , Konstantinos Moustakas This is my paper

Pith reviewed 2026-05-23 02:28 UTC · model grok-4.3

classification 💻 cs.GR cs.LG
keywords fatigue modelingphysics-informed neural networksmotion synthesisjoint torqueshuman animationmuscle fatiguebiomechanical simulation
0
0 comments X

The pith

Fatigue-PINN adapts a muscle fatigue model inside a physics-informed network to adjust maximum joint torques and synthesize realistic fatigued human motions.

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

The paper develops Fatigue-PINN to generate human movements that reflect the effects of fatigue. It embeds an established fatigue model into a neural network so that maximum torques at each joint decrease according to fatigue state. This produces joint-specific adjustments that keep motion continuous and physically plausible. The resulting system converts between joint angles and torques in one pipeline, making it usable with existing angle-based animation tools. The authors show that the generated motions align with observations from actual human fatigue experiments.

Core claim

By embedding an adaptation of the Three-Compartment Controller model inside a Physics-Informed Neural Network, Fatigue-PINN computes fatigue-driven reductions in maximum joint torques, applies joint-specific fatigue parameters to align motion across frames, and supplies an end-to-end mapping between joint angles and torques that yields smooth, experimentally consistent fatigued animations for open-type movements.

What carries the argument

The PINN adaptation of the Three-Compartment Controller model, which calculates time-varying reductions in maximum joint torques from fatigue state and supplies per-joint parameters that enforce smooth parametric alignment during motion synthesis.

Load-bearing premise

The adapted fatigue model inside the network correctly predicts how perceived fatigue lowers the maximum torque each joint can produce, and those torque changes produce motions that match real human behavior.

What would settle it

Motion-capture recordings of the same open movements performed by human subjects before and after controlled fatigue induction, compared directly against the torques and trajectories produced by Fatigue-PINN.

Figures

Figures reproduced from arXiv: 2502.19056 by Iliana Loi, Konstantinos Moustakas.

Figure 1
Figure 1. Figure 1: General overview of our Fatigue-PINN framework. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: 3CC. Figure reproduced from [24] dMR dt = −C(t) + R ∗ MF (1) dMA dt = C(t) − F ∗ MA (2) dMF dt = F ∗ MA − R ∗ MF (3) C(t) =    LD ∗ (T L − MA) if MA<T L and MR>(T L−MA) LD ∗ MR if MA<T L and MR≤(T L−MA) LR ∗ (T L − MA) if MA≥T L (4) The values of F and R are specific for every joint, as can be found in the literature [24] and depend on the task the muscles perform. A greater ratio between parameters F… view at source ↗
Figure 3
Figure 3. Figure 3: BiLSTM Inverse/Forward Dynamics Model. The feedback loop of [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The impact of different levels of fatigue on punching (Frame 290, i.e. the moment of hit), throwing (Frame 240, at throw), and waving (Frame 100) [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Effects of fatigue on punching motion. At frame [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Effects of fatigue on throwing motion. Frame [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 8
Figure 8. Figure 8: Effects of fatigue at waving motion. The starting and midwave points [PITH_FULL_IMAGE:figures/full_fig_p008_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: The impact of fatigue on lumbar bending and rotation, as well as elbow flexion, shoulder flexion, and shoulder adduction angles of the right arm [PITH_FULL_IMAGE:figures/full_fig_p009_9.png] view at source ↗
read the original abstract

Fatigue modeling is essential for motion synthesis tasks to model human motions under fatigued conditions and biomechanical engineering applications, such as investigating the variations in movement patterns and posture due to fatigue, defining injury risk mitigation and prevention strategies, formulating fatigue minimization schemes, and creating improved ergonomic designs. Nevertheless, employing datadriven methods for synthesizing the impact of fatigue on motion, receives little to no attention in the literature. In this work, we present Fatigue-PINN, a deep learning framework based on Physics-Informed Neural Networks, for modeling fatigued human movements, while providing joint-specific fatigue configurations for adaptation and mitigation of motion artifacts on a joint level, resulting in more smooth, hence physicallyplausible animations. To account for muscle fatigue, we simulate the fatigue-induced fluctuations in the maximum exerted joint torques by leveraging a PINN adaptation of the Three-Compartment Controller model to exploit physics-domain knowledge for improving accuracy. This model also introduces parametric motion alignment with respect to joint-specific fatigue, hence avoiding sharp frame transitions. Our results indicate that Fatigue-PINN accurately simulates the effects of externally perceived fatigue on open-type human movements being consistent with findings from real-world experimental fatigue studies. Since fatigue is incorporated in torque space, Fatigue-PINN provides an end-to-end encoder-decoder-like architecture, to ensure transforming joint angles to joint torques and vice-versa, thus, being compatible with motion synthesis frameworks operating on joint angles.

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

3 major / 3 minor

Summary. The manuscript presents Fatigue-PINN, a physics-informed neural network framework for fatigue-driven motion modulation and synthesis. It adapts the Three-Compartment Controller model inside a PINN to simulate fatigue-induced fluctuations in maximum joint torques, introduces joint-specific fatigue parameters for parametric alignment, and claims an end-to-end encoder-decoder architecture that transforms between joint angles and torques while producing animations consistent with real-world experimental fatigue studies.

Significance. If the adaptation rigorously enforces the original compartment ODE dynamics and the consistency claims are supported by quantitative validation, the work could provide a useful bridge between fatigue biomechanics and neural motion synthesis pipelines in computer graphics. The joint-specific configurability and torque-space integration are potentially valuable for downstream applications such as ergonomic design and injury-risk modeling.

major comments (3)
  1. [Abstract] The central claim that Fatigue-PINN 'accurately simulates the effects of externally perceived fatigue' and is 'consistent with findings from real-world experimental fatigue studies' is load-bearing, yet the abstract and model description provide no quantitative metrics (e.g., torque error, motion similarity scores, or statistical comparisons) or explicit residual terms enforcing the three coupled ODEs of the Three-Compartment Controller; without these, it is impossible to verify that the PINN outputs respect the original fatigue dynamics rather than a heuristic surrogate.
  2. [§3] §3 (model adaptation): the description of the 'PINN adaptation of the Three-Compartment Controller' does not specify the form of the physics loss (i.e., whether it includes explicit residuals on the active/fatigued/resting state rate equations or only boundary conditions on maximum torque); if the ODE residuals are absent or approximated, the torque-modulation claim reduces to a data-driven fit and loses the physics guarantees invoked in the significance statement.
  3. [§4] The assertion of 'parametric motion alignment with respect to joint-specific fatigue' and avoidance of 'sharp frame transitions' is presented without an ablation that isolates the contribution of the compartment model versus standard smoothness regularizers; this leaves open whether the observed smoothness is attributable to the claimed physics or to generic network training choices.
minor comments (3)
  1. [§3] Notation for the compartment states (active, fatigued, resting) and the fatigue rate parameters should be defined explicitly with symbols before their first use in equations.
  2. [Figures] Figure captions should include the specific fatigue levels or joint configurations shown, and axis labels should indicate units (e.g., torque in Nm).
  3. [Introduction] The manuscript should cite the original Three-Compartment Controller reference and any prior PINN adaptations of fatigue models to clarify the precise novelty of the adaptation.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed comments. We address each major comment below and indicate the revisions that will be incorporated into the next version of the manuscript.

read point-by-point responses

  1. Referee: [Abstract] The central claim that Fatigue-PINN 'accurately simulates the effects of externally perceived fatigue' and is 'consistent with findings from real-world experimental fatigue studies' is load-bearing, yet the abstract and model description provide no quantitative metrics (e.g., torque error, motion similarity scores, or statistical comparisons) or explicit residual terms enforcing the three coupled ODEs of the Three-Compartment Controller; without these, it is impossible to verify that the PINN outputs respect the original fatigue dynamics rather than a heuristic surrogate.

    Authors: We agree that the abstract would benefit from explicit quantitative support. In the revised manuscript we will add concrete metrics (torque RMSE, motion similarity scores, and statistical comparisons against experimental fatigue data) and will explicitly reference the residual terms on the three coupled ODEs so that readers can directly verify enforcement of the compartment dynamics. revision: yes

  2. Referee: [§3] §3 (model adaptation): the description of the 'PINN adaptation of the Three-Compartment Controller' does not specify the form of the physics loss (i.e., whether it includes explicit residuals on the active/fatigued/resting state rate equations or only boundary conditions on maximum torque); if the ODE residuals are absent or approximated, the torque-modulation claim reduces to a data-driven fit and loses the physics guarantees invoked in the significance statement.

    Authors: The physics loss does contain explicit residuals on the three state-rate ODEs of the Three-Compartment Controller (in addition to boundary conditions on maximum torque). We will revise §3 to state the precise form of the physics-loss term, including the weighted ODE residuals, thereby making the enforcement of the original dynamics fully transparent. revision: yes

  3. Referee: [§4] The assertion of 'parametric motion alignment with respect to joint-specific fatigue' and avoidance of 'sharp frame transitions' is presented without an ablation that isolates the contribution of the compartment model versus standard smoothness regularizers; this leaves open whether the observed smoothness is attributable to the claimed physics or to generic network training choices.

    Authors: We will add a targeted ablation study that compares the full Fatigue-PINN (with joint-specific compartment parameters) against an otherwise identical network that uses only generic smoothness regularizers. This will isolate the contribution of the physics-based fatigue model to the observed smoothness and parametric alignment. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper adapts the existing Three-Compartment Controller model inside a PINN framework to modulate maximum joint torques under fatigue. The abstract and context provide no equations, no fitted-parameter-to-prediction reductions, and no load-bearing self-citations that collapse the central claim to its own inputs by construction. The derivation therefore remains self-contained against the external compartment model and experimental benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim depends on the domain assumption of the fatigue model and the introduction of joint-specific parameters whose values are not specified as fitted or derived.

free parameters (1)
  • joint-specific fatigue configurations
    Parameters introduced for adaptation and mitigation of motion artifacts on a joint level.
axioms (1)
  • domain assumption The Three-Compartment Controller model can be adapted to simulate fatigue-induced fluctuations in maximum exerted joint torques
    Used to account for muscle fatigue in the PINN framework.

pith-pipeline@v0.9.0 · 5781 in / 1189 out tokens · 42867 ms · 2026-05-23T02:28:43.958910+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

74 extracted references · 74 canonical work pages · 2 internal anchors

  1. [1]

    Proximal and distal muscle fatigue differ- entially affect movement coordination,

    J. Cowley and D. Gates, “Proximal and distal muscle fatigue differ- entially affect movement coordination,” PLOS ONE , vol. 12, no. 2, p. e0172835, 02 2017

    work page 2017

  2. [2]

    Effects of different fatigue locations on upper body kinematics and inter-joint coordination in a repetitive pointing task,

    C. Yang, S. Leitkam, and J. N. C ˆot´e, “Effects of different fatigue locations on upper body kinematics and inter-joint coordination in a repetitive pointing task,” PLoS One, vol. 14, no. 12, 12 2019

    work page 2019

  3. [3]

    Fusing accelerometry with videography to monitor the effect of fatigue on punching performance in elite boxers,

    N. Haralabidis, D. J. Saxby, C. Pizzolato, L. Needham, D. Cazzola, and C. Minahan, “Fusing accelerometry with videography to monitor the effect of fatigue on punching performance in elite boxers,” Sensors, vol. 20, no. 20, p. 5749, 2020

    work page 2020

  4. [4]

    Impact of specialized fatigue and backhand smash on the ankle biomechanics of female badminton players,

    Z. He, G. Liu, B. Zhang, B. Ye, and H. Zhu, “Impact of specialized fatigue and backhand smash on the ankle biomechanics of female badminton players,” Scientific Reports, vol. 14, p. 10282, 05 2024

    work page 2024

  5. [5]

    Machine learning for musculoskeletal modeling of upper extremity,

    R. Sharma, A. Dasgupta, R. Cheng, C. Mishra, and V . H. Nagaraja, “Machine learning for musculoskeletal modeling of upper extremity,” IEEE Sensors Journal , vol. 22, no. 19, pp. 18 684–18 697, 2022

    work page 2022

  6. [6]

    Estimation of lower limb joint moments based on the inverse dynamics approach: a compar- ison of machine learning algorithms for rapid estimation,

    M. Mansour, K. Serbest, M. Kutlu, and M. c ¸illi, “Estimation of lower limb joint moments based on the inverse dynamics approach: a compar- ison of machine learning algorithms for rapid estimation,” Medical & biological engineering & computing , vol. 61, no. 8, 08 2023

    work page 2023

  7. [7]

    Maximum voluntary joint torque as a function of joint angle and angular velocity: Model de- velopment and application to the lower limb,

    D. Anderson, M. Madigan, and M. Nussbaum, “Maximum voluntary joint torque as a function of joint angle and angular velocity: Model de- velopment and application to the lower limb,” Journal of biomechanics, vol. 40, no. 14, pp. 3105–13, 02 2007

    work page 2007

  8. [8]

    Multi-action knee contact force prediction by domain adaptation,

    I. Loi, E. I. Zacharaki, and K. Moustakas, “Multi-action knee contact force prediction by domain adaptation,” IEEE Transactions on Neural Systems and Rehabilitation Engineering , vol. 32, pp. 122–132, 2024

    work page 2024

  9. [9]

    Prediction of medial knee contact force using multisource fusion recurrent neural network and transfer learning,

    J. Zou, X. Zhang, Y . Zhang, and Z. Jin, “Prediction of medial knee contact force using multisource fusion recurrent neural network and transfer learning,” Medical & Biological Engineering & Computing , vol. 62, no. 6, 01 2024

    work page 2024

  10. [10]

    Lower limb joint torque prediction using long short-term memory network and gaussian process regression,

    M. Wang, Z. Chen, H. Zhan, J. Zhang, X. Wu, D. Jiang, and Q. Guo, “Lower limb joint torque prediction using long short-term memory network and gaussian process regression,”Sensors, vol. 23, no. 23, 2023

    work page 2023

  11. [11]

    Knee muscle force estimating model using machine learning approach,

    A. Sohane and R. Agarwal, “Knee muscle force estimating model using machine learning approach,” The Computer Journal , vol. 65, no. 5, pp. 1167–1177, 2022

    work page 2022

  12. [12]

    Physics-informed deep learning for musculoskeletal modeling: Predicting muscle forces and joint kinematics from surface emg,

    J. Zhang, Y . Zhao, F. Shone, Z. Li, A. F. Frangi, S. Q. Xie, and Z.-Q. Zhang, “Physics-informed deep learning for musculoskeletal modeling: Predicting muscle forces and joint kinematics from surface emg,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 484–493, 2023

    work page 2023

  13. [13]

    Physics-informed deep learning for muscle force prediction with unlabeled semg signals,

    S. Ma, J. Zhang, C. Shi, P. Di, I. D. Robertson, and Z.-Q. Zhang, “Physics-informed deep learning for muscle force prediction with unlabeled semg signals,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 32, pp. 1246–1256, 2024

    work page 2024

  14. [14]

    Pimnet: Physics-infused neural network for human motion prediction,

    Z. Zhang, Y . Zhu, R. Rai, and D. Doermann, “Pimnet: Physics-infused neural network for human motion prediction,” IEEE Robotics and Automation Letters, vol. 7, no. 4, pp. 8949–8955, 2022

    work page 2022

  15. [15]

    The influence of fatigue, recovery, and environmental factors on the body stability of construction workers,

    D. Jo and H. Kim, “The influence of fatigue, recovery, and environmental factors on the body stability of construction workers,” Sensors, vol. 24, no. 11, 2024

    work page 2024

  16. [16]

    Reflex locomotion therapy for balance, gait, and fatigue rehabilitation in subjects with multiple sclerosis,

    M. Carratal ´a-Tejada, A. Cuesta-G ´omez, R. Ortiz-Guti ´errez, F. Molina- Rueda, L. Luna-Oliva, and J. C. Miangolarra-Page, “Reflex locomotion therapy for balance, gait, and fatigue rehabilitation in subjects with multiple sclerosis,” Journal of Clinical Medicine , vol. 11, no. 3, 2022

    work page 2022

  17. [17]

    Pacing and decision making in sport and exercise: the roles of perception and action in the regulation of exercise intensity,

    B. L. Smits, G.-J. Pepping, and F. J. Hettinga, “Pacing and decision making in sport and exercise: the roles of perception and action in the regulation of exercise intensity,” Sports Medicine , vol. 44, no. 6, pp. 763–775, 2014

    work page 2014

  18. [18]

    Human behaviours associated with dominance in elite amateur boxing bouts: A comparison of winners and losers under the ten point must system,

    E. C. Dunn, C. E. Humberstone, K. F. Iredale, D. T. Martin, and A. J. Blazevich, “Human behaviours associated with dominance in elite amateur boxing bouts: A comparison of winners and losers under the ten point must system,” PLoS One, vol. 12, no. 12, p. e0188675, 2017

    work page 2017

  19. [19]

    The manifestations of fatigue in amateur boxing performance,

    E. C. Dunn, “The manifestations of fatigue in amateur boxing performance,” 2019. [Online]. Available: Retrievedfromhttps://ro.ecu. edu.au/theses/2195

    work page 2019

  20. [20]

    Interventions to mitigate fatigue induced by physical work: A systematic review of research quality and levels of evidence for intervention efficacy,

    L. Lu, F. M. Megahed, and L. A. Cavuoto, “Interventions to mitigate fatigue induced by physical work: A systematic review of research quality and levels of evidence for intervention efficacy,”Human Factors: The Journal of the Human Factors and Ergonomics Society , vol. 63, no. 1, pp. 151–191, 2021

    work page 2021

  21. [21]

    Manufacturing operator ergonomics: A conceptual digital twin approach to detect biomechanical fatigue,

    A. Sharotry, J. A. Jimenez, F. A. M. Mediavilla, D. Wierschem, R. M. Koldenhoven, and D. Valles, “Manufacturing operator ergonomics: A conceptual digital twin approach to detect biomechanical fatigue,” IEEE Access, vol. 10, pp. 12 774–12 791, 2022

    work page 2022

  22. [22]

    A dynamical model of muscle activation, fatigue, and recovery,

    J. Z. Liu, R. W. Brown, and G. H. Yue, “A dynamical model of muscle activation, fatigue, and recovery,” Biophysical journal , vol. 82, no. 5, pp. 2344–2359, 2002

    work page 2002

  23. [23]

    A theoretical approach for modeling peripheral muscle fatigue and recovery,

    T. Xia and L. A. Frey Law, “A theoretical approach for modeling peripheral muscle fatigue and recovery,” Journal of Biomechanics , vol. 41, no. 14, pp. 3046–3052, 2008

    work page 2008

  24. [24]

    A three-compartment muscle fatigue model accurately predicts joint-specific maximum en- durance times for sustained isometric tasks,

    L. A. Frey Law, J. M. Looft, and J. Heitsman, “A three-compartment muscle fatigue model accurately predicts joint-specific maximum en- durance times for sustained isometric tasks,” Journal of Biomechanics , vol. 45, no. 10, pp. 1803–8, 2012

    work page 2012

  25. [25]

    Machine learning approaches for 3d motion synthesis and musculoskeletal dynamics estimation: A survey,

    I. Loi, E. I. Zacharaki, and K. Moustakas, “Machine learning approaches for 3d motion synthesis and musculoskeletal dynamics estimation: A survey,” IEEE transactions on visualization and computer graphics , vol. 30, no. 8, 08 2023

    work page 2023

  26. [26]

    Rhythm is a dancer: Music-driven motion syn- thesis with global structure,

    A. Aristidou, A. Yiannakides, K. Aberman, D. Cohen-Or, A. Shamir, and Y . Chrysanthou, “Rhythm is a dancer: Music-driven motion syn- thesis with global structure,” IEEE Transactions on Visualization and Computer Graphics, pp. 1–1, 2021

    work page 2021

  27. [27]

    Neural state machine for character-scene interactions,

    S. Starke, H. Zhang, T. Komura, and J. Saito, “Neural state machine for character-scene interactions,” ACM Transactions on Graphics , vol. 38, no. 6, p. 14, 2019

    work page 2019

  28. [28]

    Local motion phases for learning multi-contact character movements,

    S. Starke, Y . Zhao, T. Komura, and K. Zaman, “Local motion phases for learning multi-contact character movements,” ACM Transactions on Graphics, vol. 39, no. 4, p. 14, 2020. 12

    work page 2020

  29. [29]

    Neural animation layering for synthesizing martial arts movements,

    S. Starke, Y . Zhao, F. Zinno, and T. Komura, “Neural animation layering for synthesizing martial arts movements,” ACM Transactions on Graphics, vol. 40, no. 4, p. 16, 2021

    work page 2021

  30. [30]

    3d human motion synthesis based on convolutional neural network,

    D. Zhou, X. Feng, P. Yi, X. Yang, Q. Zhang, X. Wei, and D. Yang, “3d human motion synthesis based on convolutional neural network,” IEEE Access, vol. 7, pp. 66 325–66 335, 2019

    work page 2019

  31. [31]

    Gan-based reactive motion synthesis with class-aware discriminators for human-human interaction,

    Q. Men, H. Shum, E. Ho, and H. Leung, “Gan-based reactive motion synthesis with class-aware discriminators for human-human interaction,” 2021

    work page 2021

  32. [32]

    Underpressure: Deep learning for foot contact detection, ground reaction force estimation and footskate cleanup,

    L. Mourot, L. Hoyet, F. L. Clerc, and P. Hellier, “Underpressure: Deep learning for foot contact detection, ground reaction force estimation and footskate cleanup,” 2022. [Online]. Available: https: //arxiv.org/abs/2208.04598

    work page arXiv 2022

  33. [33]

    A unified 3d human motion synthesis model via conditional variational auto-encoder,

    Y . Cai, Y . Wang, Y . Zhu, T.-J. Cham, J. Cai, J. Yuan, J. Liu, C. Zheng, S. Yan, H. Ding, X. Shen, D. Liu, and N. Thalmann, “A unified 3d human motion synthesis model via conditional variational auto-encoder,” in 2021 IEEE/CVF International Conference on Computer Vision (ICCV) , 2021, pp. 11 625–11 635

    work page 2021

  34. [34]

    A unified framework for multimodal, multi-part human motion synthesis,

    Z. Zhou, Y . Wan, and B. Wang, “A unified framework for multimodal, multi-part human motion synthesis,” 2023. [Online]. Available: https://arxiv.org/abs/2311.16471

    work page arXiv 2023

  35. [35]

    A two-part transformer network for controllable motion synthesis,

    S. Hou, H. Tao, H. Bao, and W. Xu, “A two-part transformer network for controllable motion synthesis,” IEEE Transactions on Visualization and Computer Graphics , vol. 30, no. 8, pp. 5047–5062, 2024

    work page 2024

  36. [36]

    Dynamic motion transition: A hybrid data- driven and model-driven method for human pose transitions,

    Z. Chai and H. Qin, “Dynamic motion transition: A hybrid data- driven and model-driven method for human pose transitions,” IEEE Transactions on Visualization and Computer Graphics , pp. 1–14, 2024

    work page 2024

  37. [37]

    Single motion diffusion,

    S. Raab, I. Leibovitch, G. Tevet, M. Arar, A. H. Bermano, and D. Cohen-Or, “Single motion diffusion,” 2023. [Online]. Available: https://arxiv.org/abs/2302.05905

    work page arXiv 2023

  38. [38]

    Listen, denoise, action! audio-driven motion synthesis with diffusion models,

    S. Alexanderson, R. Nagy, J. Beskow, and G. E. Henter, “Listen, denoise, action! audio-driven motion synthesis with diffusion models,” ACM Trans. Graph., vol. 42, no. 4, 2023

    work page 2023

  39. [39]

    Mofusion: A framework for denoising-diffusion-based motion synthesis,

    R. Dabral, M. H. Mughal, V . Golyanik, and C. Theobalt, “Mofusion: A framework for denoising-diffusion-based motion synthesis,” in Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2023, pp. 9760–9770

    work page 2023

  40. [40]

    Guess: Gradually enriching synthesis for text-driven human motion generation,

    X. Gao, Y . Yang, Z. Xie, S. Du, Z. Sun, and Y . Wu, “Guess: Gradually enriching synthesis for text-driven human motion generation,” IEEE Transactions on Visualization and Computer Graphics , vol. 30, no. 12, pp. 7518–7530, 2024

    work page 2024

  41. [41]

    Learning a family of motor skills from a single motion clip,

    S. Lee, S. Lee, Y . Lee, and J. Lee, “Learning a family of motor skills from a single motion clip,” ACM Transactions on Graphics , vol. 40, no. 4, pp. 1–13, 2021

    work page 2021

  42. [42]

    Universal humanoid motion representations for physics-based control,

    Z. Luo, J. Cao, J. Merel, A. Winkler, J. Huang, K. Kitani, and W. Xu, “Universal humanoid motion representations for physics-based control,”

    work page

  43. [43]

    Available: https://arxiv.org/abs/2310.04582

    [Online]. Available: https://arxiv.org/abs/2310.04582

    work page arXiv

  44. [44]

    Force: Dataset and method for intuitive physics guided human-object interaction,

    X. Zhang, B. L. Bhatnagar, S. Starke, I. Petrov, V . Guzov, H. Dhamo, E. P ´erez-Pellitero, and G. Pons-Moll, “Force: Dataset and method for intuitive physics guided human-object interaction,” 2024. [Online]. Available: https://arxiv.org/abs/2403.11237

    work page arXiv 2024

  45. [45]

    Discovering fatigued movements for virtual character animation,

    N. Cheema, R. Xu, N. H. Kim, P. H ¨am¨al¨ainen, V . Golyanik, M. Haber- mann, C. Theobalt, and P. Slusallek, “Discovering fatigued movements for virtual character animation,” in SIGGRAPH Asia 2023 Conference Papers. New York, NY , USA: Association for Computing Machinery, 2023

    work page 2023

  46. [46]

    Opensim: Simulating musculoskeletal dynamics and neuromuscular control to study human and animal movement,

    A. Seth, J. Hicks, T. Uchida, A. Habib, C. Dembia, J. Dunne, C. Ong, M. DeMers, A. Rajagopal, M. Millard, S. Hamner, E. Arnold, J. Yong, S. Lakshmikanth, M. Sherman, and S. Delp, “Opensim: Simulating musculoskeletal dynamics and neuromuscular control to study human and animal movement,” PLoS Computational Biology , vol. 14, no. 7, p. e1006223, 2018

    work page 2018

  47. [47]

    Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations

    M. Raissi, P. Perdikaris, and G. E. Karniadakis, “Physics informed deep learning (part I): data-driven solutions of nonlinear partial differential equations,” CoRR, vol. abs/1711.10561, 2017. [Online]. Available: http://arxiv.org/abs/1711.10561

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  48. [48]

    Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations,

    M. Raissi, P. Perdikaris, and G. Karniadakis, “Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations,” Journal of Computational Physics, vol. 378, pp. 686–707, 2019

    work page 2019

  49. [49]

    nn- pinns: Non-newtonian physics-informed neural networks for complex fluid modeling,

    M. Mahmoudabadbozchelou, G. E. Karniadakis, and S. Jamali, “nn- pinns: Non-newtonian physics-informed neural networks for complex fluid modeling,” Soft Matter, vol. 18, no. 1, pp. 172–185, 2022

    work page 2022

  50. [50]

    Physics-informed deep-learning applications to experimental fluid mechanics,

    H. Eivazi, Y . Wang, and R. Vinuesa, “Physics-informed deep-learning applications to experimental fluid mechanics,” Measurement science and technology, vol. 35, no. 7, p. 075303, 2024

    work page 2024

  51. [51]

    Hybrid quantum physics-informed neural networks for simulating computational fluid dynamics in complex shapes,

    A. Sedykh, M. Podapaka, A. Sagingalieva, K. Pinto, M. Pflitsch, and A. Melnikov, “Hybrid quantum physics-informed neural networks for simulating computational fluid dynamics in complex shapes,” Machine Learning: Science and Technology, vol. 5, no. 2, p. 025045, May 2024. [Online]. Available: http://dx.doi.org/10.1088/2632-2153/ad43b2

    work page doi:10.1088/2632-2153/ad43b2 2024

  52. [52]

    Quantum physics-informed neural networks,

    C. Trahan, M. Loveland, and S. Dent, “Quantum physics-informed neural networks,” Entropy, vol. 26, no. 8, p. 649, 2024

    work page 2024

  53. [53]

    A multi-resolution physics- informed recurrent neural network: formulation and application to musculoskeletal systems,

    K. Taneja, X. He, Q. He, and J.-S. Chen, “A multi-resolution physics- informed recurrent neural network: formulation and application to musculoskeletal systems,” Computational Mechanics, vol. 73, no. 5, pp. 1125–1145, 2024

    work page 2024

  54. [54]

    Predicting multi-joint kinematics of the upper limb from emg signals across varied loads with a physics-informed neural network,

    R. Kumar, S. P. Muthukrishnan, L. Kumar, and S. Roy, “Predicting multi-joint kinematics of the upper limb from emg signals across varied loads with a physics-informed neural network,” 2023. [Online]. Available: https://arxiv.org/abs/2312.09418

    work page arXiv 2023

  55. [55]

    Boosting personalized musculoskeletal modeling with physics-informed knowledge transfer,

    J. Zhang, Y . Zhao, T. Bao, Z. Li, K. Qian, A. Frangi, S. Xie, and E. Zhang, “Boosting personalized musculoskeletal modeling with physics-informed knowledge transfer,”IEEE Transactions on Instrumen- tation and Measurement , vol. PP, pp. 1–1, 11 2022

    work page 2022

  56. [56]

    semg-driven physics-informed gated recurrent networks for modeling upper limb multi-joint movement dynamics,

    R. Kumar, A. Gupta, S. P. Muthukrishnan, L. Kumar, and S. Roy, “semg-driven physics-informed gated recurrent networks for modeling upper limb multi-joint movement dynamics,” 2024. [Online]. Available: https://arxiv.org/abs/2408.16599

    work page arXiv 2024

  57. [57]

    Toward robust and effi- cient musculoskeletal modeling using distributed physics-informed deep learning,

    J. Zhang, Z. Ruan, Q. Li, and Z.-Q. Zhang, “Toward robust and effi- cient musculoskeletal modeling using distributed physics-informed deep learning,” IEEE Transactions on Instrumentation and Measurement , vol. 72, pp. 1–11, 2023

    work page 2023

  58. [58]

    Learning phrase representations using rnn encoder-decoder for statistical machine translation,

    K. Cho, B. van Merri ¨enboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y . Bengio, “Learning phrase representations using rnn encoder-decoder for statistical machine translation,” in Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2014, pp. 1724–1734

    work page 2014

  59. [59]

    Chollet, Deep Learning with Python

    F. Chollet, Deep Learning with Python . New York: Manning Publica- tions, 2017

    work page 2017

  60. [60]

    Authoring multi-actor behaviors in crowds with diverse personalities,

    M. Kapadia, A. Shoulson, F. Durupinar, and N. I. Badler, “Authoring multi-actor behaviors in crowds with diverse personalities,” Modeling, Simulation and Visual Analysis of Crowds: A Multidisciplinary Perspec- tive, pp. 147–180, 2013

    work page 2013

  61. [61]

    Full-body musculoskeletal model for muscle-driven sim- ulation of human gait,

    A. Rajagopal, C. L. Dembia, M. S. DeMers, D. D. Delp, J. L. Hicks, and S. L. Delp, “Full-body musculoskeletal model for muscle-driven sim- ulation of human gait,” IEEE transactions on biomedical engineering , vol. 63, no. 10, pp. 2068–2079, 2016

    work page 2068

  62. [62]

    Predicting mid-air interaction movements and fatigue using deep reinforcement learning,

    N. Cheema, L. A. Frey-Law, K. Naderi, J. Lehtinen, P. Slusallek, and P. H¨am¨al¨ainen, “Predicting mid-air interaction movements and fatigue using deep reinforcement learning,” in Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems , 2020, pp. 1–13

    work page 2020

  63. [63]

    Muscle fatigue modelling: Solving for fatigue and recovery parameter values using fewer maximum effort assessments,

    L. A. Frey-Law, M. Schaffer, and F. K. Urban III, “Muscle fatigue modelling: Solving for fatigue and recovery parameter values using fewer maximum effort assessments,” International journal of industrial ergonomics, vol. 82, p. 103104, 2021

    work page 2021

  64. [64]

    Estimation of critical torque using intermittent isometric maximal voluntary contractions of the quadriceps in humans,

    M. Burnley, “Estimation of critical torque using intermittent isometric maximal voluntary contractions of the quadriceps in humans,” Journal of Applied Physiology , vol. 106, no. 3, pp. 975–983, 2009

    work page 2009

  65. [65]

    Chollet et al

    F. Chollet et al. (2015) Keras. [Online]. Available: https://github.com/ fchollet/keras

    work page 2015

  66. [66]

    Adam: A method for stochastic optimization,

    D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,”

    work page

  67. [67]

    Adam: A Method for Stochastic Optimization

    [Online]. Available: https://arxiv.org/abs/1412.6980

    work page internal anchor Pith review Pith/arXiv arXiv

  68. [68]

    Bidirectional recurrent neural networks,

    M. Schuster and K. Paliwal, “Bidirectional recurrent neural networks,” IEEE Transactions on Signal Processing , vol. 45, no. 11, pp. 2673 – 2681, 1997

    work page 1997

  69. [69]

    A Deep Learning Approach for Human Activities Recognition From Multimodal Sensing Devices,

    I. Ihianle, A. Nwajana, S. Ebenuwa, R. Otuka, K. Owa, and M. Orisatoki, “A Deep Learning Approach for Human Activities Recognition From Multimodal Sensing Devices,” IEEE Access , vol. 8, pp. 179 028– 179 038, 2020

    work page 2020

  70. [70]

    Deep bidirectional and unidirectional lstm recurrent neural network for network-wide traffic speed prediction,

    Z. Cui, R. Ke, Z. Pu, and Y . Wang, “Deep bidirectional and unidirectional lstm recurrent neural network for network-wide traffic speed prediction,” 2018. [Online]. Available: https://arxiv.org/abs/1801. 02143

    work page 2018

  71. [71]

    A history of the unity game engine,

    J. K. Haas, “A history of the unity game engine,” 2014

    work page 2014

  72. [72]

    Posture-movement changes following repetitive motion-induced shoulder muscle fatigue,

    J. R. Fuller, K. V . Lomond, J. Fung, and J. N. C ˆot´e, “Posture-movement changes following repetitive motion-induced shoulder muscle fatigue,” Journal of Electromyography and Kinesiology, vol. 19, no. 6, pp. 1043– 1052, 2009

    work page 2009

  73. [73]

    The impact of fatigue on the kinematics of collegiate baseball pitchers,

    W. J. Grantham, I. R. Byram, M. C. Meadows, and C. S. Ah- mad, “The impact of fatigue on the kinematics of collegiate baseball pitchers,” Orthopaedic journal of sports medicine , vol. 2, no. 6, p. 2325967114537032, 2014. 13

    work page 2014

  74. [74]

    Combining recurrent neural networks and adversarial training for human motion synthesis and control,

    Z. Wang, J. Chai, and S. Xia, “Combining recurrent neural networks and adversarial training for human motion synthesis and control,” IEEE Transactions on Visualization and Computer Graphics , vol. 27, no. 1, pp. 14–28, 2019

    work page 2019