pith. sign in

arxiv: 2504.03468 · v2 · submitted 2025-04-04 · 💻 cs.CV

D-Garment: Physically Grounded Latent Diffusion for Dynamic Garment Deformations

Antoine Dumoulin , Adnane Boukhayma , Laurence Boissieux , Bharath Bhushan Damodaran , Pierre Hellier , Stefanie Wuhrer This is my paper

Pith reviewed 2026-05-22 21:05 UTC · model grok-4.3

classification 💻 cs.CV
keywords garment deformationlatent diffusionphysical material propertiesdynamic wrinklescloth simulation3D meshbody motion conditioning
0
0 comments X

The pith

A latent diffusion model in 2D parameter space generates 3D garment deformations conditioned on physical cloth material properties.

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

The paper introduces a method to deform 3D garment meshes using body shape, motion, and physical cloth properties. It trains a generative model on data produced by a physics-based simulator so that the outputs respect physical constraints such as strain and curvature. The model operates as a template-specific latent diffusion process in 2D parameter space, allowing independent conditioning on global and local geometry from body and material inputs. This setup improves handling of loose clothing and dynamic wrinkles compared with prior pose-only approaches and supports fitting to observed 3D point clouds. Evaluation on both simulated and captured data shows gains in shape similarity and physical validity metrics.

Core claim

D-Garment is a learning-based approach trained on physics-simulator data that learns garment deformations conditioned by physical material properties. This conditioning enables modeling of loose cloth geometry, especially large deformations and dynamic wrinkles driven by body motion. The model is realized as a template-specific latent diffusion model in 2D parameter space that conditions global and local geometry with body and cloth material information and can be fitted to vision-sensor observations such as 3D point clouds.

What carries the argument

Template-specific latent diffusion model operating in 2D parameter space, conditioned on body shape, motion, and physical cloth material properties.

If this is right

  • The model produces garment shapes that better respect physical validity metrics such as strain and curvature than prior methods.
  • It supports efficient fitting of the generative prior to 3D point-cloud observations from vision sensors.
  • Dynamic wrinkles and large deformations in loose clothing become modellable when material properties are provided as conditioning input.
  • Quantitative improvements appear in both shape-similarity and physically inspired error measures on simulated and captured test data.

Where Pith is reading between the lines

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

  • The same conditioning strategy could be tested on other classes of deformable objects whose motion is governed by material parameters.
  • Real captured sequences could be used to fine-tune the simulator-trained model and reduce any domain gap between simulation and observation.
  • Interactive design tools could allow users to vary material parameters and immediately see the resulting change in dynamic garment behavior.

Load-bearing premise

Training data produced by the physics-based simulator accurately captures real-world cloth dynamics and material behavior for the garments and motions considered.

What would settle it

A direct comparison of model outputs against real multi-view captured garment sequences that reveals consistent mismatches in measured strain, curvature, or wrinkle formation for given material properties and body motions would falsify the physical-grounding claim.

Figures

Figures reproduced from arXiv: 2504.03468 by Adnane Boukhayma, Antoine Dumoulin, Bharath Bhushan Damodaran, Laurence Boissieux, Pierre Hellier, Stefanie Wuhrer.

Figure 1
Figure 1. Figure 1: We introduce a latent diffusion model that allows to gen [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: D-Garment generates garment deformations conditioned on body shape, motion and cloth material. It builds upon a 2D latent diffusion model (Sec. 3.2) to learn how to deform a template in uv-space (Sec. 3.1). 3D mesh vertex displacement from template is parameterized by the uv displacement map, and our model is trained on it along with the conditional inputs. At inference, our model generates the deformed ga… view at source ↗
Figure 3
Figure 3. Figure 3: Qualitative comparison of two garment simulations [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Examples generated by varying one cloth material parameter at a time. The model provides control over bending, stretching and [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Examples generated by varying one of the principal com [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: Qualitative example of the fitting application shown [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗
read the original abstract

We present a method to dynamically deform 3D garments, in the form of a 3D polygon mesh, based on body shape, motion, and physical cloth material properties. Considering physical cloth properties allows to learn a physically grounded model, with the advantage of being more accurate in terms of physically inspired metrics such as strain or curvature. Existing work studies pose-dependent garment modeling to generate garment deformations from example data, and possibly data-driven dynamic cloth simulation to generate realistic garments in motion. We propose D-Garment, a learning-based approach trained on new data generated with a physics-based simulator. Compared to prior work, our 3D generative model learns garment deformations conditioned by physical material properties, which allows to model loose cloth geometry, especially for large deformations and dynamic wrinkles driven by body motion. Furthermore, the model can be efficiently fitted to observations captured using vision sensors such as 3D point clouds. We leverage the capability of diffusion models to learn flexible and powerful generative priors by modeling the 3D garment in a 2D parameter space independently from the mesh resolution. This representation allows to learn a template-specific latent diffusion model. This allows to condition global and local geometry with body and cloth material information. We quantitatively and qualitatively evaluate D-Garment on both simulations and data captured with a multi-view acquisition platform. Compared to recent baselines, our method is more realistic and accurate in terms of shape similarity and physical validity metrics. Code and data are available for research purposes at https://dumoulina.github.io/d-garment/

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 / 2 minor

Summary. The manuscript presents D-Garment, a template-specific latent diffusion model that generates dynamic 3D garment deformations (as polygon meshes) conditioned on body shape, pose/motion, and physical cloth material properties. The model is trained on data produced by a physics-based simulator, operates in a 2D parameter space to decouple from mesh resolution, and is shown to fit to real 3D point clouds; quantitative results claim improvements over baselines in both shape similarity and physical validity metrics (strain, curvature) on simulated and multi-view captured data.

Significance. If the central claims hold, the work would advance data-driven garment modeling by explicitly conditioning generative priors on material properties, improving realism for loose clothing and motion-driven wrinkles. The choice of a resolution-independent 2D latent diffusion representation and the public release of code and data are practical strengths that aid reproducibility and extension.

major comments (2)
  1. [Evaluation] Evaluation section (results on simulations and captured data): Physical validity metrics (strain, curvature) are reported exclusively on held-out data generated by the same physics-based simulator used to create the training set. Real captured data from the multi-view platform is used only for shape similarity, so the claimed superiority on physically inspired metrics remains dependent on simulator fidelity rather than providing an independent test of real-world physical grounding.
  2. [Method] Method and abstract: The central claim that material-property conditioning enables better modeling of 'large deformations and dynamic wrinkles' lacks a dedicated ablation that isolates the contribution of the material parameters from other conditioning signals (body shape/motion) or from the diffusion architecture itself; without this, it is difficult to attribute the reported gains specifically to the physical grounding.
minor comments (2)
  1. [Abstract] Abstract: The phrase 'new data generated with a physics-based simulator' should specify the simulator name, the exact material parameter ranges sampled, and the garment types/motions covered to allow readers to assess coverage.
  2. [Method] The manuscript would benefit from a clearer statement of the precise loss terms or conditioning mechanisms used to inject material properties into the latent diffusion process.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback. We address each major comment below and describe the revisions we will make to strengthen the manuscript.

read point-by-point responses

  1. Referee: [Evaluation] Evaluation section (results on simulations and captured data): Physical validity metrics (strain, curvature) are reported exclusively on held-out data generated by the same physics-based simulator used to create the training set. Real captured data from the multi-view platform is used only for shape similarity, so the claimed superiority on physically inspired metrics remains dependent on simulator fidelity rather than providing an independent test of real-world physical grounding.

    Authors: We thank the referee for this observation. Physical validity metrics are evaluated on held-out simulated data because the physics-based simulator supplies precise ground-truth strain and curvature values, enabling quantitative assessment of physical behavior that is difficult to obtain directly from real captures. On the real multi-view captured data we report shape similarity to demonstrate fitting to observed point clouds, and we include qualitative results showing realistic motion-driven wrinkles consistent with the learned physical priors. In the revision we will add an explicit discussion paragraph clarifying the simulator's role in validating physical metrics and how the real-data fitting results support generalization of these properties. revision: partial

  2. Referee: [Method] Method and abstract: The central claim that material-property conditioning enables better modeling of 'large deformations and dynamic wrinkles' lacks a dedicated ablation that isolates the contribution of the material parameters from other conditioning signals (body shape/motion) or from the diffusion architecture itself; without this, it is difficult to attribute the reported gains specifically to the physical grounding.

    Authors: We agree that a dedicated ablation isolating material-parameter conditioning would strengthen attribution of the observed gains. While the current results compare against baselines that lack material conditioning and show corresponding improvements in physical metrics, we will add a new ablation experiment in the revised manuscript. This will train and evaluate a variant of D-Garment without material inputs on the same shape and physical-validity metrics, allowing direct quantification of the material parameters' contribution to large deformations and dynamic wrinkles. revision: yes

Circularity Check

0 steps flagged

Evaluation of physical metrics on held-out simulator data is standard but limits real-world grounding claim

full rationale

The paper trains a latent diffusion model on garment meshes generated by an external physics-based simulator and reports physical validity metrics (strain, curvature) on held-out simulation data from the same simulator. This follows standard ML train/test protocol on external data rather than fitting parameters directly to the reported metrics or defining the target via the model itself. Real captured multi-view data is used separately for shape similarity evaluation, and comparisons are made to independent baselines. No equations or steps reduce the central generative claim to a self-referential fit or self-citation chain by construction. The minor elevation to score 2 reflects that physical-accuracy superiority is demonstrated only within the simulator distribution, leaving simulator-to-real fidelity as an untested external assumption rather than a circularity in the derivation.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The approach rests on the domain assumption that the external physics simulator supplies faithful training targets for strain, curvature, and material behavior; no free parameters or invented entities are stated in the abstract.

axioms (1)
  • domain assumption Physics-based simulator produces training data that accurately reflect real cloth dynamics and material properties
    All quantitative claims rest on comparison against data generated by this simulator

pith-pipeline@v0.9.0 · 5836 in / 1211 out tokens · 42488 ms · 2026-05-22T21:05:16.159747+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

61 extracted references · 61 canonical work pages · 1 internal anchor

  1. [1]

    A comparative study of garment draping techniques

    Prerana Achar, Mayank Patel, Anushka Mulik, Neha Ka- tre, Stevina Dias, and Chirag Raman. A comparative study of garment draping techniques. arXiv preprint arXiv:2405.11056, 2024. 2

    work page arXiv 2024

  2. [2]

    4DHumanOutfit: a multi- subject 4D dataset of human motion sequences in varying outfits exhibiting large displacements

    Matthieu Armando, Laurence Boissieux, Edmond Boyer, Jean-S´ebastien Franco, Martin Humenberger, Christophe Legras, Vincent Leroy, Mathieu Marsot, Julien Pansiot, Sergi Pujades, Rim Rekik Dit Nekhili, Gr ´egory Rogez, Anilku- mar Swamy, and Stefanie Wuhrer. 4DHumanOutfit: a multi- subject 4D dataset of human motion sequences in varying outfits exhibiting l...

    work page 2023

  3. [3]

    Large steps in cloth sim- ulation

    David Baraff and Andrew Witkin. Large steps in cloth sim- ulation. In SIGGRAPH, pages 767–778, 1998. 2, 4

    work page 1998

  4. [4]

    Neu- ral cloth simulation

    Hugo Bertiche, Meysam Madadi, and Sergio Escalera. Neu- ral cloth simulation. ACM Transactions on Graphics, 41(6): 1–14, 2022. 2

    work page 2022

  5. [5]

    Chat- garment: Garment estimation, generation and editing via large language models

    Siyuan Bian, Chenghao Xu, Yuliang Xiu, Artur Grigorev, Zhen Liu, Cewu Lu, Michael J Black, and Yao Feng. Chat- garment: Garment estimation, generation and editing via large language models. arXiv preprint arXiv:2412.17811 ,

    work page arXiv

  6. [6]

    Projective dynamics: fusing constraint projections for fast simulation.ACM Transactions on Graph- ics, 33(4), 2014

    Sofien Bouaziz, Sebastian Martin, Tiantian Liu, Ladislav Ka- van, and Mark Pauly. Projective dynamics: fusing constraint projections for fast simulation.ACM Transactions on Graph- ics, 33(4), 2014. 2

    work page 2014

  7. [7]

    Gaps: Geometry-aware, physics-based, self-supervised neural gar- ment draping

    Ruochen Chen, Liming Chen, and Shaifali Parashar. Gaps: Geometry-aware, physics-based, self-supervised neural gar- ment draping. In International Conference on 3D Vision , pages 116–125, 2024. 2, 3

    work page 2024

  8. [8]

    Aleny `a, Gerard Pons- Moll, and Francesc Moreno-Noguer

    Enric Corona, Albert Pumarola, G. Aleny `a, Gerard Pons- Moll, and Francesc Moreno-Noguer. Smplicit: Topology- aware generative model for clothed people. Conference on Computer Vision and Pattern Recognition , pages 11870– 11880, 2021. 2

    work page 2021

  9. [9]

    Drapenet: Garment generation and self- supervised draping

    Luca De Luigi, Ren Li, Beno ˆıt Guillard, Mathieu Salzmann, and Pascal Fua. Drapenet: Garment generation and self- supervised draping. In Conference on Computer Vision and Pattern Recognition, pages 1451–1460, 2023. 2, 3, 4

    work page 2023

  10. [10]

    Scaling recti- fied flow transformers for high-resolution image synthesis

    Patrick Esser, Sumith Kulal, Andreas Blattmann, Rahim Entezari, Jonas M ¨uller, Harry Saini, Yam Levi, Dominik Lorenz, Axel Sauer, Frederic Boesel, et al. Scaling recti- fied flow transformers for high-resolution image synthesis. In Forty-first international conference on machine learning,

    work page

  11. [11]

    Black, and Otmar Hilliges

    Artur Grigorev, Michael J. Black, and Otmar Hilliges. HOOD: hierarchical graphs for generalized modelling of clothing dynamics. In Conference on Computer Vision and Pattern Recognition, 2023. 2, 3, 6, 7

    work page 2023

  12. [12]

    Geome- try images

    Xianfeng Gu, Steven J Gortler, and Hugues Hoppe. Geome- try images. In Conference on Computer Graphics and Inter- active Techniques, pages 355–361, 2002. 4 9

    work page 2002

  13. [13]

    Diffusion shape prior for wrinkle-accurate cloth registration

    Jingfan Guo, Fabian Prada, Donglai Xiang, Javier Romero, Chenglei Wu, Hyun Soo Park, Takaaki Shiratori, and Shun- suke Saito. Diffusion shape prior for wrinkle-accurate cloth registration. In International Conference on 3D Vision, page 790–799, 2024. 5

    work page 2024

  14. [14]

    Dresscode: Autoregressively sewing and gen- erating garments from text guidance

    Kai He, Kaixin Yao, Qixuan Zhang, Jingyi Yu, Lingjie Liu, and Lan Xu. Dresscode: Autoregressively sewing and gen- erating garments from text guidance. ACM Transactions on Graphics, 43(4), 2024. 2

    work page 2024

  15. [15]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. Advances in Neural Information Processing Systems, 33:6840–6851, 2020. 2, 4, 5

    work page 2020

  16. [16]

    Adam: A Method for Stochastic Optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 ,

    work page internal anchor Pith review Pith/arXiv arXiv

  17. [17]

    A generative multi-resolution pyramid and normal-conditioning 3d cloth draping

    Hunor Laczk ´o, Meysam Madadi, Sergio Escalera, and Jordi Gonzalez. A generative multi-resolution pyramid and normal-conditioning 3d cloth draping. In Winter Confer- ence on Applications of Computer Vision, pages 8709–8718,

    work page

  18. [18]

    Brown, and Lau- rence Boissieux

    Jie Li, Gilles Daviet, Rahul Narain, Florence Bertails- Descoubes, Matthew Overby, George E. Brown, and Lau- rence Boissieux. An implicit frictional contact solver for adaptive cloth simulation. ACM Transactions on Graphics, 37(4), 2018. 2

    work page 2018

  19. [19]

    Kaufman, Vladimir G

    Minchen Li, Danny M. Kaufman, Vladimir G. Kim, Justin Solomon, and Alla Sheffer. Optcuts: joint optimization of surface cuts and parameterization. ACM Transactions on Graphics, 37(6), 2018. 6

    work page 2018

  20. [20]

    Common diffusion noise schedules and sample steps are flawed

    Shanchuan Lin, Bingchen Liu, Jiashi Li, and Xiao Yang. Common diffusion noise schedules and sample steps are flawed. In Winter Conference on Applications of Computer Vision, pages 5404–5411, 2024. 6

    work page 2024

  21. [21]

    Matthew Loper, Naureen Mahmood, Javier Romero, Gerard Pons-Moll, and Michael J. Black. Smpl: a skinned multi- person linear model. ACM Transactions on Graphics, 34(6),

    work page

  22. [22]

    Decoupled weight de- cay regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight de- cay regularization. In International Conference on Learning Representations, 2019. 6

    work page 2019

  23. [23]

    Dpm-solver: A fast ode solver for diffusion probabilistic model sampling in around 10 step s

    Cheng Lu, Yuhao Zhou, Fan Bao, Jianfei Chen, Chongxuan Li, and Jun Zhu. Dpm-solver++: Fast solver for guided sam- pling of diffusion probabilistic models. arXiv:2206.00927,

    work page arXiv

  24. [24]

    Projective Dynamics with Dry Fric- tional Contact

    Micka ¨el Ly, Jean Jouve, Laurence Boissieux, and Florence Bertails-Descoubes. Projective Dynamics with Dry Fric- tional Contact. ACM Transactions on Graphics, 39(4), 2020. 2, 5

    work page 2020

  25. [25]

    Qianli Ma, Jinlong Yang, Anurag Ranjan, Sergi Pujades, Gerard Pons-Moll, Siyu Tang, and Michael J. Black. Learn- ing to Dress 3D People in Generative Clothing. InComputer Vision and Pattern Recognition, 2020. 3

    work page 2020

  26. [26]

    Neural point-based shape modeling of humans in challeng- ing clothing

    Qianli Ma, Jinlong Yang, Michael J Black, and Siyu Tang. Neural point-based shape modeling of humans in challeng- ing clothing. In International Conference on 3D Vision , pages 679–689, 2022. 3

    work page 2022

  27. [27]

    Troje, Ger- ard Pons-Moll, and Michael J

    Naureen Mahmood, Nima Ghorbani, Nikolaus F. Troje, Ger- ard Pons-Moll, and Michael J. Black. AMASS: Archive of motion capture as surface shapes. International Conference on Computer Vision, pages 5441–5450, 2019. 2, 5

    work page 2019

  28. [28]

    Representing motion as a sequence of latent primitives, a flexible approach for human motion modelling

    Mathieu Marsot, Stefanie Wuhrer, Jean-Sebastien Franco, and Anne H´el`ene Olivier. Representing motion as a sequence of latent primitives, a flexible approach for human motion modelling. arXiv preprint arXiv:2206.13142, 2022. 9

    work page arXiv 2022

  29. [29]

    COAP: Compositional articu- lated occupancy of people

    Marko Mihajlovic, Shunsuke Saito, Aayush Bansal, Michael Zollhoefer, and Siyu Tang. COAP: Compositional articu- lated occupancy of people. In Conference on Computer Vi- sion and Pattern Recognition, 2022. 2

    work page 2022

  30. [30]

    Aipparel: A large multimodal generative model for digital garments

    Kiyohiro Nakayama, Jan Ackermann, Timur Levent Kes- dogan, Yang Zheng, Maria Korosteleva, Olga Sorkine- Hornung, Leonidas Guibas, Guandao Yang, and Gordon Wetzstein. Aipparel: A large multimodal generative model for digital garments. In Conference on Computer Vision and Pattern Recognition (CVPR), 2025. 2

    work page 2025

  31. [31]

    Rahul Narain, Armin Samii, and James F. O’Brien. Adaptive anisotropic remeshing for cloth simulation. ACM Transac- tions on Graphics, 31:1 – 10, 2012. 2

    work page 2012

  32. [32]

    Laplacian mesh optimization

    Andrew Nealen, Takeo Igarashi, Olga Sorkine, and Marc Alexa. Laplacian mesh optimization. In International Con- ference on Computer Graphics and Interactive Techniques , page 381–389, 2006. 5

    work page 2006

  33. [33]

    Shape and pose space deformation for subject specific animation

    Alexandro Neophytou and Adrian Hilton. Shape and pose space deformation for subject specific animation. In Inter- national Conference on 3D Vision, pages 334–341, 2013. 2

    work page 2013

  34. [34]

    Tailornet: Predicting clothing in 3d as a function of hu- man pose, shape and garment style.Conference on Computer Vision and Pattern Recognition, pages 7363–7373, 2020

    Chaitanya Patel, Zhouyingcheng Liao, and Gerard Pons- Moll. Tailornet: Predicting clothing in 3d as a function of hu- man pose, shape and garment style.Conference on Computer Vision and Pattern Recognition, pages 7363–7373, 2020. 3, 4

    work page 2020

  35. [35]

    Building statistical shape spaces for 3d human modeling

    Leonid Pishchulin, Stefanie Wuhrer, Thomas Helten, Chris- tian Theobalt, and Bernt Schiele. Building statistical shape spaces for 3d human modeling. Pattern Recognition, 67: 276–286, 2017. 2

    work page 2017

  36. [36]

    State of the art on diffusion models for visual computing

    Ryan Po, Wang Yifan, Vladislav Golyanik, Kfir Aberman, Jonathan T Barron, Amit Bermano, Eric Chan, Tali Dekel, Aleksander Holynski, Angjoo Kanazawa, et al. State of the art on diffusion models for visual computing. In Computer Graphics Forum, page e15063. Wiley Online Library, 2024. 4

    work page 2024

  37. [37]

    SDXL: Improving latent diffusion models for high-resolution image synthesis

    Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas M ¨uller, Joe Penna, and Robin Rombach. SDXL: Improving latent diffusion models for high-resolution image synthesis. In International Con- ference on Learning Representations, 2024. 6

    work page 2024

  38. [38]

    Deformation constraints in a mass-spring model to describe rigid cloth behaviour

    Xavier Provot. Deformation constraints in a mass-spring model to describe rigid cloth behaviour. In Graphics Inter- face, pages 147–147, 1995. 2

    work page 1995

  39. [39]

    High-resolution image syn- thesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bjorn Ommer. High-resolution image syn- thesis with latent diffusion models. In Conference on Com- puter Vision and Pattern Recognition, 2022. 2, 4, 5

    work page 2022

  40. [40]

    Physical validation of sim- ulators in computer graphics: A new framework dedicated to 10 slender elastic structures and frictional contact

    Victor Romero, Micka ¨el Ly, Abdullah Haroon Rasheed, Rapha¨el Charrondi`ere, Arnaud Lazarus, S´ebastien Neukirch, and Florence Bertails-Descoubes. Physical validation of sim- ulators in computer graphics: A new framework dedicated to 10 slender elastic structures and frictional contact. ACM Trans- actions on Graphics, 40(4):1–19, 2021. 3

    work page 2021

  41. [41]

    U- net: Convolutional networks for biomedical image segmen- tation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U- net: Convolutional networks for biomedical image segmen- tation. In Medical image computing and computer-assisted intervention, pages 234–241, 2015. 6

    work page 2015

  42. [42]

    Self-supervised collision handling via generative 3d garment models for virtual try-on

    Igor Santesteban, Nils Thuerey, Miguel A Otaduy, and Dan Casas. Self-supervised collision handling via generative 3d garment models for virtual try-on. In Conference on Com- puter Vision and Pattern Recognition , pages 11763–11773,

    work page

  43. [43]

    Snug: Self-supervised neural dynamic garments

    Igor Santesteban, Miguel A Otaduy, and Dan Casas. Snug: Self-supervised neural dynamic garments. In Conference on Computer Vision and Pattern Recognition , pages 8140– 8150, 2022. 2, 3

    work page 2022

  44. [44]

    Garment3DGen: 3d garment stylization and texture generation

    Nikolaos Sarafianos, Tuur Stuyck, Xiaoyu Xiang, Yilei Li, Jovan Popovic, and Rakesh Ranjan. Garment3DGen: 3d garment stylization and texture generation. In International Conference on 3D Vision, 2025. 2

    work page 2025

  45. [45]

    Gan-based gar- ment generation using sewing pattern images

    Yu Shen, Junbang Liang, and Ming C Lin. Gan-based gar- ment generation using sewing pattern images. In European Conference on Computer Vision, pages 225–247, 2020. 2, 3

    work page 2020

  46. [46]

    Gener- ating diverse clothed 3d human animations via a generative model

    Min Shi, Wenke Feng, Lin Gao, and Dengming Zhu. Gener- ating diverse clothed 3d human animations via a generative model. Computational Visual Media, 10(2):261–277, 2024. 3

    work page 2024

  47. [47]

    Denois- ing diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denois- ing diffusion implicit models. In International Conference on Learning Representations, 2021. 5

    work page 2021

  48. [48]

    GenSim: Unsupervised generic garment simulator

    Lokender Tiwari, Brojeshwar Bhowmick, and Sanjana Sinha. GenSim: Unsupervised generic garment simulator. In Conference on Computer Vision and Pattern Recognition, pages 4169–4178, 2023. 2

    work page 2023

  49. [49]

    Millimetric Human Surface Capture in Minutes

    Briac Toussaint, Laurence Boissieux, Diego Thomas, Ed- mond Boyer, and Franco Jean-S´ebastien. Millimetric Human Surface Capture in Minutes. In SIGGRAPH Asia, 2024. 9

    work page 2024

  50. [50]

    Capturing hands in action using discriminative salient points and physics sim- ulation

    Dimitrios Tzionas, Luca Ballan, Abhilash Srikantha, Pablo Aponte, Marc Pollefeys, and Juergen Gall. Capturing hands in action using discriminative salient points and physics sim- ulation. International Journal of Computer Vision , 118(2): 172–193, 2016. 6

    work page 2016

  51. [51]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017. 6

    work page 2017

  52. [52]

    Diffused- wrinkles: A diffusion-based model for data-driven garment animation

    Raquel Vidaurre, Elena Garces, and Dan Casas. Diffused- wrinkles: A diffusion-based model for data-driven garment animation. In British Machine Vision Conference, 2024. 2, 3, 4

    work page 2024

  53. [53]

    Diffusers: State-of-the-art diffu- sion models

    Patrick von Platen, Suraj Patil, Anton Lozhkov, Pedro Cuenca, Nathan Lambert, Kashif Rasul, Mishig Davaadorj, Dhruv Nair, Sayak Paul, William Berman, Yiyi Xu, Steven Liu, and Thomas Wolf. Diffusers: State-of-the-art diffu- sion models. https://github.com/huggingface/ diffusers, 2022. 6

    work page 2022

  54. [54]

    End-to-end diffusion latent optimization improves classifier guidance

    Bram Wallace, Akash Gokul, Stefano Ermon, and Nikhil Naik. End-to-end diffusion latent optimization improves classifier guidance. In International Conference on Com- puter Vision, pages 7280–7290, 2023. 5

    work page 2023

  55. [55]

    Wang, Duygu Ceylan, Jovan Popovi ´c, and Niloy J

    Tuanfeng Y . Wang, Duygu Ceylan, Jovan Popovi ´c, and Niloy J. Mitra. Learning a shared shape space for multi- modal garment design. ACM Transactions on Graphics, 37 (6), 2018. 2

    work page 2018

  56. [56]

    GHUM & GHUML: Generative 3d human shape and artic- ulated pose models

    Hongyi Xu, Eduard Gabriel Bazavan, Andrei Zanfir, Bill Freeman, Rahul Sukthankar, and Cristian Sminchisescu. GHUM & GHUML: Generative 3d human shape and artic- ulated pose models. In Conference on Computer Vision and Pattern Recognition, pages 6184–6193, 2020. 2

    work page 2020

  57. [57]

    Estimation of human body shape in motion with wide clothing

    Jinlong Yang, Jean-S ´ebastien Franco, Franck H ´etroy- Wheeler, and Stefanie Wuhrer. Estimation of human body shape in motion with wide clothing. In European Confer- ence on Computer Vision, pages 439–454, 2016. 5, 9

    work page 2016

  58. [58]

    Analyzing clothing layer de- formation statistics of 3d human motions

    Jinlong Yang, Jean-Sebastien Franco, Franck Hetroy- Wheeler, and Stefanie Wuhrer. Analyzing clothing layer de- formation statistics of 3d human motions. In European Con- ference on Computer Vision, 2018. 3, 7

    work page 2018

  59. [59]

    Meng Zhang, Duygu Ceylan, and Niloy J. Mitra. Motion guided deep dynamic 3d garments. ACM Transactions on Graphics, 41(6), 2022. 3, 6, 7

    work page 2022

  60. [60]

    Design2Cloth: 3d cloth generation from 2d masks

    Jiali Zheng, Rolandos Alexandros Potamias, and Stefanos Zafeiriou. Design2Cloth: 3d cloth generation from 2d masks. In Conference on Computer Vision and Pattern Recognition, pages 1748–1758, 2024. 2

    work page 2024

  61. [61]

    On the continuity of rotation representations in neural networks

    Yi Zhou, Connelly Barnes, Jingwan Lu, Jimei Yang, and Hao Li. On the continuity of rotation representations in neural networks. In Conference on Computer Vision and Pattern Recognition, pages 5745–5753, 2019. 6 11

    work page 2019