EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

GraphNeuralCloth: A Graph‐Neural‐Network‐Based Framework for Non‐Skinning Cloth Simulation

Yingqi Li, Xiaomei Wang, Xiaoqi Song, Chenchen Fan, Bohao Zhu, Jianan Peng, Qixin Liang, James Lam, Kam‐Yim Sze, Ka‐Wai Kwok

Advanced Intelligent Systems · 2026-04 · 8 (4)

DOI: 10.1002/aisy.202501120 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advanced Intelligent Systems, online erschienen am 10.02.2026. Verfasst von Yingqi Li und 9 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cloth simulation has a wide range of applications, spanning from virtual animation (e.g., video games and movies) to robotic manipulation (e.g., automatic cloth folding). Conventional physics‐based methods involve computationally exhaustive iterations, making real‐time simulation challenging. Deep learning provides an alternative for fast and realistic cloth simulation but most of these learning models are still trained using virtual data only. In this article, we present a cloth motion capture system and a point‐cloud‐to‐mesh processing method to support the prediction of real‐world fabric deformation. We also propose GraphNeuralCloth, a graph‐neural‐network (GNN)‐based framework, capable of estimating the cloth morphology change in real time. Our framework has been evaluated on three datasets, showcasing its performance not only in avatar try‐on (≈4.2 mm in nodal distance error using Santesteban's dataset [ 1 ] ), but also in real fabric free falling (≈2.3 mm using fabric draping dataset). In addition, our network model can accurately predict changes of real‐world cloth morphology (≈8.55 mm using cloth motion dataset). The ablation study demonstrates that the key components of our model, namely spiral graph convolution and identity connection, contribute to faster training convergence and reduced inference time (≈1000 fps).

Zitieren

APA

Li, Y., Wang, X., Song, X., Fan, C., Zhu, B., Peng, J., Liang, Q., Lam, J., Sze, K., & Kwok, K. (2026). GraphNeuralCloth: A Graph‐Neural‐Network‐Based Framework for Non‐Skinning Cloth Simulation. Advanced Intelligent Systems, 8(4). https://doi.org/10.1002/aisy.202501120

Vancouver

Li Y, Wang X, Song X, Fan C, Zhu B, Peng J, et al. GraphNeuralCloth: A Graph‐Neural‐Network‐Based Framework for Non‐Skinning Cloth Simulation. Advanced Intelligent Systems. 2026;8(4). doi:10.1002/aisy.202501120

BibTeX
@article{Li2026,
  author = {Li, Yingqi and Wang, Xiaomei and Song, Xiaoqi and Fan, Chenchen and Zhu, Bohao and Peng, Jianan and Liang, Qixin and Lam, James and Sze, Kam‐Yim and Kwok, Ka‐Wai},
  title = {GraphNeuralCloth: A Graph‐Neural‐Network‐Based Framework for Non‐Skinning Cloth Simulation},
  journal = {Advanced Intelligent Systems},
  year = {2026},
  volume = {8},
  number = {4},
  doi = {10.1002/aisy.202501120}
}
Kostenlos starten →