Pith. sign in

REVIEW 2 major objections 4 minor 41 references

Wearing A Coat: Dual-Arm Robot-Assisted Dressing with Differentiable Clothing Simulation

T0 review · 2 major / 4 minor · reviewed 2026-07-14 · grok-4.5

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Dual-arm robots can fully put coats on people by simulating cloth contacts in real time under partial-wear constraints.

desk verdict Working dual-arm coat dressing system with a usable large-step explicit differentiable cloth sim; physical claims rest on single-run demos rather than quantified reliability. read the letter →

arxiv 2607.10999 v1 pith:A44TVZXE submitted 2026-07-13 cs.RO

Yiming Liu , Lijun Han , Hesheng Wang This is my paper
classification cs.RO
keywords robot-assisteddressingdifferentiableclothingsimulationhuman-robotinteractiondual-armcontrolclothmodelpredictivecontactconstraints
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Putting a coat on someone with a robot is hard because once one sleeve is on, the fabric is tightly constrained by contact with the arm, and the two sleeves cannot move independently. Most prior methods treat the garment as loose segments and stop after the first sleeve. This paper shows that a fast, differentiable cloth simulator—built on an explicit iteration that adds carefully chosen high-order bias terms for stability at large time steps—can resolve the full garment state under dry-friction contact. That state feeds a multi-stage planner that dresses one sleeve, adjusts the body pose, then dresses the second sleeve while predicting human motion. A simpler constrained local model supplies high-frequency corrections so the robot can react at 10 Hz. Physical experiments with different coats, standing and sitting poses, and both passive and active people confirm that both sleeves reach the shoulders, which would expand practical assistive dressing for people with limited mobility.

What carries the argument

The position-velocity-decoupled explicit iteration with high-order bias (Eqs. 7–12) plus dry-friction contact update, which yields both forward garment states and analytic Jacobians for gradient-based global control; a linearized local dressing model with active-set constrained DDP then supplies the real-time corrections.

What would settle it

In a physical trial with active human motion and a previously unseen coat, measure whether both progress scalars reach exactly 2 while peak simulated inner forces stay below the safety threshold used in the cost; failure of either metric under otherwise identical conditions would refute the claim that the simulation-plus-control loop is sufficient.

Watch

Extended reading notes

Core claim

An explicit iterative differentiable clothing simulator that intentionally injects high-order bias terms remains stable and accurate enough under large time steps to supply contact-resolved garment states; those states, combined with multi-stage objective-driven dual-arm model-predictive control and a constrained local compensator, let a robot complete full dual-sleeve coat dressing for varied poses, garments, and human motion patterns.

Load-bearing premise

The high-order bias terms keep the fast explicit simulation accurate enough for contact forces and sleeve tracking at the chosen large steps without artificial damping that would invalidate the dressing predictions.

Editorial extensions

If this is right

  • Robots can finish the second sleeve of a conventional coat even when the wearer’s initial arm span exceeds shoulder width.
  • Assistive systems can switch online between passive compliance and active human cooperation without garment-specific pre-models.
  • The same contact-aware multi-stage decomposition becomes available for other constrained soft-object tasks such as jacket removal or layered dressing.
  • Real-time differentiable cloth simulation at 0.01 s steps is now practical for closed-loop dual-arm MPC on commodity hardware.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same high-order-bias explicit scheme could accelerate differentiable simulation for other frictional soft-body contacts outside dressing, such as blanket covering or bag packing.
  • Multi-stage sleeve sequencing may transfer directly to lower-body garments once a comparable local compensator is written for pants legs.
  • Because the global controller already predicts human intent, adding force-torque feedback at the grippers would likely tighten the safety margins without changing the architecture.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper presents a dual-arm robot-assisted coat dressing system that couples a novel explicit, position-velocity-decoupled differentiable cloth simulator (with intentional high-order bias terms for large-step stability and dry-friction contact) to a multi-stage model-predictive controller. The multi-stage objective (Eq. 32) sequences first-sleeve insertion, arm adjustment under garment constraint, and second-sleeve donning; a constrained local linear model solved by an active-set CDDP variant supplies high-frequency compensation. Perception uses GarmentNets-style reconstruction plus RTMPose and a diffusion-based human-motion predictor. Validation comprises simulation comparisons against Projective Dynamics (Figs. 5–6, Table I), a single-sleeve gradient-control demo (Figs. 7–8), and physical dummy/human trials across garments, standing/sitting poses, and passive/active motion (Figs. 10–14), with progress scalars s_i reaching 2 and force ablations.

Significance. If the claims hold, the work supplies a practical route to full dual-sleeve coat dressing under contact constraints—an open problem that prior segment-based or single-sleeve methods leave unsolved. The explicit high-order-bias simulator is a concrete engineering contribution that demonstrably improves real-time performance relative to PD while remaining differentiable, and the multi-stage + local-compensation architecture is a reusable pattern for other constrained deformable-object HRI tasks. Physical success on both dummy and human subjects with two garments and both passive and active intent is a non-trivial system-level result for assistive robotics.

major comments (2)
  1. Sec. VII-C and Figs. 10–14: the central claim that the integrated system “successfully completes dual-sleeve coat dressing under contact constraints for varied poses, garments, and both passive and active human motion” rests exclusively on single-run qualitative sequences and progress-scalar/force traces. No multi-trial success rates, inter-subject variance, confidence intervals, or systematic failure-mode analysis are reported. Without quantified reliability, the efficacy of the multi-stage objective (Eq. 32) and the constrained local compensator (Alg. 2) under realistic contact and intent variation remains unproven; at minimum a success-rate table over repeated trials with controlled pose/garment/intent factors is required.
  2. Eqs. (7)–(12) and Conclusion: the intentionally introduced high-order bias terms are acknowledged by the authors to produce “more pronounced energy dissipation” and to render the method “less suitable for highly dynamic scenarios.” Dressing involves rapid contact transitions and sleeve sliding; the paper never quantifies how much artificial damping distorts contact-force predictions or sleeve-state tracking at the operating steps (h = 0.01 s local, h = 1 s global). A short sensitivity study (e.g., energy residual or force error versus h against a fine-step PD reference under sleeve-contact conditions) is needed to confirm that the bias does not invalidate the contact-constrained control claims.
minor comments (4)
  1. Table I and Figs. 5–6: computational times for the placement task rise sharply with contact; a brief complexity discussion (O(n_c^{3} + n^{2})) already appears later, but an explicit statement of typical n_c under dressing conditions would help readers assess real-time margins.
  2. Notation: the same symbol ϕ is overloaded for multiple distinct thresholds (ϕ_p, ϕ_c, ϕ_f, ϕ_x, ϕ_v); a single table of all free parameters and their numerical values used in experiments would improve reproducibility.
  3. Fig. 1 pipeline diagram is dense; the multi-stage strategy block could be enlarged or split so that the three phases (first sleeve / arm adjust / second sleeve) are visually distinct.
  4. Related-work discussion of DiffCloth / DiffPD is accurate but could more clearly state which contact and friction gradients are newly derived versus reused.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the explicit high-order-bias simulator, multi-stage costs, and local compensator are engineering constructions whose success metric (s_i reaching 2) is independent of free parameters and is checked against external PD baselines plus physical hardware.

full rationale

The derivation chain begins from a second-order Taylor expansion of nodal position/velocity (Eq. 1), augments it with intentionally introduced high-order bias terms that recover a semi-definite mass-matrix compensation (Eqs. 7–12), and obtains an explicit fixed-point iteration whose spectral radius is shown <1 by Rayleigh-quotient argument (Eq. 14). Contact forces are obtained by a Signorini–Coulomb projection that is algebraically independent of the later control objectives (Eqs. 15–20). Differentiability follows by direct matrix differentiation of the same update (Eqs. 21–27). The multi-stage cost J_task (Eq. 32) is a hand-designed sequence of quadratic tracking terms whose stage transitions are thresholded on free scalars ϕ_p, ϕ_c; those thresholds do not appear inside the success metric s_i=2. The local linear model and constrained DDP solver (Alg. 2) are likewise free-parameter constructions. Human-motion training data are generated from the same local model, yet the final claim is physical dual-sleeve completion under both passive and active human motion, measured by independent progress scalars and force traces (Figs. 10–14). No equation reduces a claimed prediction to a fitted input by construction, no uniqueness theorem is imported from the authors’ prior work, and the only external benchmark (Projective Dynamics) is an independent algorithm. Consequently the experimental outcome is not forced by the modeling choices.

Assumptions & free parameters 5 free parameters · 4 assumptions · 2 invented entities

The central claim rests on standard continuum-mechanics and robotics assumptions plus several free thresholds/weights chosen for the dressing stages and a new high-order bias construction whose stability is argued via spectral-radius and limiting-equilibrium analysis rather than external theorem. No new physical particles or forces are postulated; the 'invented' pieces are algorithmic.

free parameters (5)
  • time step h (local/global)
    Chosen as 0.01 s for local and 1 s for global predictive optimization; stability and accuracy claims depend on these values.
  • stage thresholds ϕ_p, ϕ_c, ϕ_f, ϕ_x, ϕ_v
    Hand-tuned scalars that switch dressing stages and enforce local feasibility; alter the multi-stage policy behavior.
  • cost weights W, Q, R, w_s
    Diagonal matrices and scalar in the MPC objectives; fitted or chosen to balance tracking, orientation, and safety.
  • mesh resolution / node count
    5 cm clustering (~1800 nodes) selected for real-time trade-off; changes contact count and force accuracy.
  • friction coefficient μ and material stiffnesses w_i
    Garment-specific inputs assumed known; contact classification and force magnitudes depend on them.
assumptions (4)
  • domain assumption Cloth potential energy decomposes into quadratic stretching/bending terms projectable onto SO(3) manifolds (Projective Dynamics style).
    Invoked in Sec. IV-A Eqs. 2–6 to obtain internal forces and their derivatives; standard in the cited PD literature.
  • domain assumption Signorini-Coulomb contact law with constant force per step and mean-velocity approximation is sufficient for dressing contacts.
    Used in Sec. IV-B Eqs. 15–19; common nonsmooth-mechanics model but approximate for sliding fabric.
  • ad hoc to paper High-order bias terms of the form h² F_x Δx / Δv keep the explicit scheme stable for large h while remaining o(h³).
    Core construction of Sec. IV-A Eqs. 7–12; justified by spectral-radius argument and limiting equilibrium but not derived from a prior theorem.
  • domain assumption Human intent is either passive (follow garment) or active (move to random targets) and can be classified by an attention discriminator + diffusion predictor.
    Sec. VI-B; enables the multi-stage planner under occlusion.
invented entities (2)
  • Position-velocity-decoupled explicit iteration with intentional high-order bias
    purpose: Enable real-time large-step differentiable cloth simulation that remains stable under high stiffness.
    Novel algorithmic construct introduced in Sec. IV; independent evidence is the PD comparison and physical dressing success, but no external formal verification.
  • Constrained local linear dressing model + active-set CDDP solver
    purpose: Provide high-frequency compensation for the slow global simulator while respecting fabric stretch and contact inequalities.
    Introduced in Sec. V-B/C and Algorithm 2; evidence is the ablation showing higher update rate and better second-sleeve success.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Wearing A Coat: Dual-Arm Robot-Assisted Dressing with Differentiable Clothing Simulation." pith.science (2026). https://pith.science/paper/A44TVZXE

@misc{pith2026260710999,
  author       = {Pith},
  title        = {Pith review of: Wearing A Coat: Dual-Arm Robot-Assisted Dressing with Differentiable Clothing Simulation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A44TVZXE}},
  note         = {Machine review of arXiv:2607.10999}
}
read the original abstract

The development of assistive robots for dressing tasks serves to augment human convenience and improve the quality of life for individuals with physical impairments. However, due to the intricate contact interactions between garments and the human limbs during dressing, most robot-assisted dressing algorithms treat clothing as an assembly of discrete segments, thereby struggling to manage the partial worn garments under contact constraints. To overcome this challenge, we propose a novel robotic dressing control algorithm that integrates realtime differentiable clothing simulation. The simulation algorithm employs explicit iterative scheme with intentionally introduced higher-order perturbations to enhance computational efficiency while maintaining stability under large time-step conditions. Through simulation, we resolve the garment state under contact constraints, which then enables a multi-phase control strategy for successful coat dressing assistance. To further improve real-time performance, we introduce a constrained local model along with its corresponding optimization solver, permitting high-frequency local compensation for the differentiable simulation based global controller. Finally, we experimentally validate our approach through both simulated and physical dressing scenarios, conclusively demonstrating its feasibility and efficacy

Figures

Figures reproduced from arXiv: 2607.10999 by the authors.

Figure 1
Figure 1. The pipeline of our proposed robotic-assisted dressing framework. The system comprises three components: human and garment perception, [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. An example of the process to solve the contact force: A rectangular [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. The control strategy designed for the dressing task. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Simulation results of our proposed algorithm with step [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: The results of draping task. Our method and the PD algo [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 8
Figure 8. Figure 8: (a) The motion trajectories of the sleeve state [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]
Figure 9
Figure 9. Figure 9: The reconstruction of the human body and garment in the point cloud. [PITH_FULL_IMAGE:figures/full_fig_p013_9.png]
Figure 10
Figure 10. Figure 10: The task phase segmentation and the corresponding sleeve states during dummy dressing. [PITH_FULL_IMAGE:figures/full_fig_p014_10.png]
Figure 11
Figure 11. Figure 11: Evolution of the sleeve state and maximum simulated inner forces [PITH_FULL_IMAGE:figures/full_fig_p014_11.png]
Figure 12
Figure 12. Figure 12: The assisted dressing outcomes for two garments in both standing and sitting poses with human passive motion. Each case is illustrated through [PITH_FULL_IMAGE:figures/full_fig_p015_12.png]
Figure 13
Figure 13. Figure 13: The assisted dressing process with active human cooperation. According to the variations in human strategy, the donning process is illustrated through [PITH_FULL_IMAGE:figures/full_fig_p015_13.png]
Figure 14
Figure 14. Figure 14: Evolution of the sleeve state and maximum simulated inner forces [PITH_FULL_IMAGE:figures/full_fig_p015_14.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

41 extracted references · 1 linked inside Pith

  1. [1]

    Physically assistive robots: A systematic review of mobile and manipulator robots that physically assist people with disabilities,

    A. Nanavati, V . Ranganeni, and M. Cakmak, “Physically assistive robots: A systematic review of mobile and manipulator robots that physically assist people with disabilities,”Annual Review of Control, Robotics, and Autonomous Systems, vol. 7, 2023

    2023

  2. [2]

    From screens to scenes: A survey of embodied ai in healthcare,

    Y . Liu, X. Cao, T. Chen, Y . Jiang, J. You, M. Wu, X. Wang, M. Feng, Y . Jin, and J. Chen, “From screens to scenes: A survey of embodied ai in healthcare,”Information Fusion, vol. 119, p. 103033, 2025

    2025

  3. [3]

    What does the person feel? Learning to infer applied forces during robot-assisted dressing,

    Z. Erickson, A. Clegg, W. Yu, G. Turk, C. K. Liu, and C. C. Kemp, “What does the person feel? Learning to infer applied forces during robot-assisted dressing,” in2017 IEEE International Conference on Robotics and Automation (ICRA). IEEE, May 2017, pp. 6058–6065

    2017

  4. [4]

    Deep Haptic Model Predictive Control for Robot-Assisted Dressing,

    Z. Erickson, H. M. Clever, G. Turk, C. K. Liu, and C. C. Kemp, “Deep Haptic Model Predictive Control for Robot-Assisted Dressing,” in2018 IEEE International Conference on Robotics and Automation (ICRA), May 2018, pp. 4437–4444

    2018

  5. [5]

    Personalized collaborative plans for robot-assisted dressing via optimization and simulation,

    A. Kapusta, Z. Erickson, H. M. Clever, W. Yu, C. K. Liu, G. Turk, and C. C. Kemp, “Personalized collaborative plans for robot-assisted dressing via optimization and simulation,”Autonomous Robots, vol. 43, no. 8, pp. 2183–2207, Dec. 2019

    2019

  6. [6]

    Do you need a Hand? – A Bimanual Robotic Dressing Assistance Scheme,

    J. Zhu, M. Gienger, G. Franzese, and J. Kober, “Do you need a Hand? – A Bimanual Robotic Dressing Assistance Scheme,”IEEE Transactions on Robotics, vol. 40, pp. 1906–1919, 2024

    1906

  7. [7]

    Iterative path optimisation for personalised dressing assistance using vision and force information,

    Y . Gao, H. J. Chang, and Y . Demiris, “Iterative path optimisation for personalised dressing assistance using vision and force information,” in2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2016, pp. 4398–4403

    2016

  8. [8]

    Cross-Domain Representation Learning for Clothes Unfolding in Robot-Assisted Dressing,

    J. Qie, Y . Gao, R. Feng, X. Wang, J. Yang, E. Dasgupta, H. J. Chang, and Y . Chang, “Cross-Domain Representation Learning for Clothes Unfolding in Robot-Assisted Dressing,” inComputer Vision – ECCV 2022 Workshops, L. Karlinsky, T. Michaeli, and K. Nishino, Eds. Springer Nature Switzerland, 2023, vol. 13806, pp. 658–671

    2022

Show all 41 references
  1. [9]

    Personalized assist-as- needed dressing assistance robot without human modeling using rowat- selverston cpg controller,

    K. Yamasaki, T. Shibata, and P. H ´enaff, “Personalized assist-as- needed dressing assistance robot without human modeling using rowat- selverston cpg controller,”Advanced Robotics, vol. 38, no. 19-20, pp. 1408–1423, 2024

    2024

  2. [10]

    Probabilistic Real-Time User Posture Tracking for Personalized Robot-Assisted Dressing,

    F. Zhang, A. Cully, and Y . Demiris, “Probabilistic Real-Time User Posture Tracking for Personalized Robot-Assisted Dressing,”IEEE Transactions on Robotics, vol. 35, no. 4, pp. 873–888, Aug. 2019

    2019

  3. [11]

    Motion Generation for Mobile Manipulators to Assist in Putting on a Jacket While Walking,

    Y . Kuribayashi, Y . Takase, and K. Yamazaki, “Motion Generation for Mobile Manipulators to Assist in Putting on a Jacket While Walking,” in2024 IEEE 20th International Conference on Automation Science and Engineering (CASE). Bari, Italy: IEEE, Aug. 2024, pp. 2229–2236

    2024

  4. [12]

    Learning adaptive dressing assistance from human demonstration,

    E. Pignat and S. Calinon, “Learning adaptive dressing assistance from human demonstration,”Robotics and Autonomous Systems, vol. 93, pp. 61–75, Jul. 2017

    2017

  5. [13]

    A framework for robotic clothing assistance by imitation learning,

    R. P. Joshi, N. Koganti, and T. Shibata, “A framework for robotic clothing assistance by imitation learning,”Advanced Robotics, vol. 33, no. 22, pp. 1156–1174, Nov. 2019

    2019

  6. [14]

    Learning to Collaborate From Simulation for Robot-Assisted Dressing,

    A. Clegg, Z. Erickson, P. Grady, G. Turk, C. C. Kemp, and C. K. Liu, “Learning to Collaborate From Simulation for Robot-Assisted Dressing,” IEEE Robotics and Automation Letters, vol. 5, no. 2, pp. 2746–2753, Apr. 2020

    2020

  7. [15]

    Learning garment manipulation policies toward robot-assisted dressing,

    F. Zhang and Y . Demiris, “Learning garment manipulation policies toward robot-assisted dressing,”Science Robotics, vol. 7, no. 65, p. eabm6010, Apr. 2022

    2022

  8. [16]

    Force-Constrained Visual Policy: Safe Robot-Assisted Dressing via Multi-Modal Sensing,

    Z. Sun, Y . Wang, D. Held, and Z. Erickson, “Force-Constrained Visual Policy: Safe Robot-Assisted Dressing via Multi-Modal Sensing,”IEEE Robotics and Automation Letters, vol. 9, no. 5, pp. 4178–4185, May 2024

    2024

  9. [17]

    Large steps in cloth simulation,

    D. Baraff and A. Witkin, “Large steps in cloth simulation,” inPro- ceedings of the 25th Annual Conference on Computer Graphics and Interactive Techniques - SIGGRAPH ’98. ACM Press, 1998, pp. 43– 54

    1998

  10. [18]

    Fast simulation of mass-spring systems,

    T. Liu, A. W. Bargteil, J. F. O’Brien, and L. Kavan, “Fast simulation of mass-spring systems,”ACM Transactions on Graphics, vol. 32, no. 6, pp. 1–7, Nov. 2013

    2013

  11. [19]

    Projective Dy- namics: Fusing Constraint Projections for Fast Simulation,

    S. Bouaziz, S. Martin, T. Liu, L. Kavan, and M. Pauly, “Projective Dy- namics: Fusing Constraint Projections for Fast Simulation,” inSeminal Graphics Papers: Pushing the Boundaries, V olume 2, 1st ed., M. C. Whitton, Ed. ACM, 2023, pp. 787–797

    2023

  12. [20]

    Admm⊇projective dynamics: fast simulation of general constitutive models,

    R. Narain, M. Overby, and G. E. Brown, “Admm⊇projective dynamics: fast simulation of general constitutive models,” inProceedings of the ACM SIGGRAPH/Eurographics Symposium on Computer Animation, ser. SCA ’16. Goslar, DEU: Eurographics Association, 2016, p. 21–28

    2016

  13. [21]

    Quasi-Newton Methods for Real- Time Simulation of Hyperelastic Materials,

    T. Liu, S. Bouaziz, and L. Kavan, “Quasi-Newton Methods for Real- Time Simulation of Hyperelastic Materials,”ACM Transactions on Graphics, vol. 36, no. 3, pp. 1–16, 2017

    2017

  14. [22]

    Position based dynamics,

    M. M ¨uller, B. Heidelberger, M. Hennix, and J. Ratcliff, “Position based dynamics,”J. Vis. Comun. Image Represent., vol. 18, no. 2, p. 109–118, Apr. 2007

    2007

  15. [23]

    Position-based simulation of continuous materials,

    J. Bender, D. Koschier, P. Charrier, and D. Weber, “Position-based simulation of continuous materials,”Computers & Graphics, vol. 44, no. 0, pp. 1 – 10, 2014

    2014

  16. [24]

    XPBD: Position-based simulation of compliant constrained dynamics,

    M. Macklin, M. M ¨uller, and N. Chentanez, “XPBD: Position-based simulation of compliant constrained dynamics,” inProceedings of the 9th International Conference on Motion in Games. Burlingame California: ACM, Oct. 2016, pp. 49–54

    2016

  17. [25]

    Garment Diffusion Models for Robot- Assisted Dressing,

    S. Kotsovolis and Y . Demiris, “Garment Diffusion Models for Robot- Assisted Dressing,”IEEE Robotics and Automation Letters, vol. 10, no. 2, pp. 1217–1224, Feb. 2025

    2025

  18. [26]

    DiffCloth: Differentiable Cloth Simulation with Dry Frictional Contact,

    Y . Li, T. Du, K. Wu, J. Xu, and W. Matusik, “DiffCloth: Differentiable Cloth Simulation with Dry Frictional Contact,”ACM Transactions on Graphics, vol. 42, no. 1, pp. 1–20, Feb. 2023

    2023

  19. [27]

    Differentiable simulation of soft multi-body systems,

    Y .-L. Qiao, J. Liang, V . Koltun, and M. Lin, “Differentiable simulation of soft multi-body systems,” inAdvances in Neural Information Processing Systems, A. Beygelzimer, Y . Dauphin, P. Liang, and J. W. Vaughan, Eds., 2021

    2021

  20. [28]

    Liang, M

    J. Liang, M. C. Lin, and V . Koltun,Differentiable cloth simulation for inverse problems. Red Hook, NY , USA: Curran Associates Inc., 2019

    2019

  21. [29]

    Diffpd: Differentiable projective dynamics,

    T. Du, K. Wu, P. Ma, S. Wah, A. Spielberg, D. Rus, and W. Ma- tusik, “Diffpd: Differentiable projective dynamics,”ACM Trans. Graph., vol. 41, no. 2, Nov. 2021

    2021

  22. [30]

    Projective dynamics with dry frictional contact,

    M. Ly, J. Jouve, L. Boissieux, and F. Bertails-Descoubes, “Projective dynamics with dry frictional contact,”ACM Trans. Graph., vol. 39, no. 4, Aug. 2020

    2020

  23. [31]

    DiffClothAI: Dif- ferentiable Cloth Simulation with Intersection-free Frictional Contact and Differentiable Two-Way Coupling with Articulated Rigid Bodies,

    X. Yu, S. Zhao, S. Luo, G. Yang, and L. Shao, “DiffClothAI: Dif- ferentiable Cloth Simulation with Intersection-free Frictional Contact and Differentiable Two-Way Coupling with Articulated Rigid Bodies,” in2023 IEEE/RSJ International Conference on Intelligent Robots and System...

    2023

  24. [32]

    Incremental potential contact: intersection-and inversion-free, large-deformation dynamics,

    M. Li, Z. Ferguson, T. Schneider, T. Langlois, D. Zorin, D. Panozzo, C. Jiang, and D. M. Kaufman, “Incremental potential contact: intersection-and inversion-free, large-deformation dynamics,”ACM Trans. Graph., vol. 39, no. 4, Aug. 2020

    2020

  25. [33]

    Brogliato,Nonsmooth Lagrangian Systems

    B. Brogliato,Nonsmooth Lagrangian Systems. Springer Cham, 2016, pp. 241–370

    2016

  26. [34]

    D. E. Kirk,Optimal control theory: an introduction. Courier Corpora- tion, 2004

    2004

  27. [35]

    Differential dynamic programming with nonlinear constraints,

    Z. Xie, C. K. Liu, and K. Hauser, “Differential dynamic programming with nonlinear constraints,” in2017 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2017, pp. 695–702

    2017

  28. [36]

    GarmentNets: Category-Level Pose Estimation for Garments via Canonical Space Shape Completion,

    C. Chi and S. Song, “GarmentNets: Category-Level Pose Estimation for Garments via Canonical Space Shape Completion,” in2021 IEEE/CVF International Conference on Computer Vision (ICCV). Montreal, QC, Canada: IEEE, Oct. 2021, pp. 3304–3313

    2021

  29. [37]

    CLOTH3D: Clothed 3D Humans,

    H. Bertiche, M. Madadi, and S. Escalera, “CLOTH3D: Clothed 3D Humans,” inComputer Vision – ECCV 2020, A. Vedaldi, H. Bischof, T. Brox, and J.-M. Frahm, Eds. Cham: Springer International Publish- ing, 2020, vol. 12365, pp. 344–359

    2020

  30. [38]

    Assistive gym: A physics simulation framework for assistive robotics,

    Z. Erickson, V . Gangaram, A. Kapusta, C. K. Liu, and C. C. Kemp, “Assistive gym: A physics simulation framework for assistive robotics,” IEEE International Conference on Robotics and Automation (ICRA), pp. 10 169–10 176, 2020

    2020

  31. [39]

    Rtmw: Real-time multi-person 2d and 3d whole-body pose estimation,

    T. Jiang, X. Xie, and Y . Li, “Rtmw: Real-time multi-person 2d and 3d whole-body pose estimation,” 2024. [Online]. Available: https://arxiv.org/abs/2407.08634 JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 17

    2024 arXiv

  32. [40]

    Leapfrog diffu- sion model for stochastic trajectory prediction,

    W. Mao, C. Xu, Q. Zhu, S. Chen, and Y . Wang, “Leapfrog diffu- sion model for stochastic trajectory prediction,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 5517–5526

    2023

  33. [41]

    Cotracker: It is better to track together,

    N. Karaev, I. Rocco, B. Graham, N. Neverova, A. Vedaldi, and C. Rup- precht, “Cotracker: It is better to track together,” inComputer Vision – ECCV 2024. Springer Nature Switzerland, 2024, pp. 18–35

    2024

Pith tools

Reviewed July 14, 2026 · model on record in the stance chip above.