EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-SA 4.0

PS-CAD: Local Geometry Guidance via Prompting and Selection for CAD Reconstruction

Bingchen Yang, Haiyong Jiang, Hao Pan, Guosheng Lin, Jun Xiao, Peter Wonka

ACM Transactions on Graphics · 30.06.2025 · 44 (3) · 1-21

DOI: 10.1145/3733595 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Graphics, online erschienen am 29.05.2025. Verfasst von Bingchen Yang und 5 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0). Gefördert von: National Natural Science Foundation of China, National Key R&D Program of China, Beijing Nova Program, Beijing Natural Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Reverse engineering CAD models from raw geometry is a classic but challenging research problem. In particular, reconstructing the CAD modeling sequence from point clouds provides great interpretability and convenience for editing. Analyzing previous work, we observed that a CAD modeling sequence represented by tokens and processed by a generative model does not have an immediate geometric interpretation. To improve upon this problem, we introduce geometric guidance into the reconstruction network. Our proposed model, PS-CAD, reconstructs the CAD modeling sequence one step at a time as illustrated in Figure 1 . At each step, we provide three forms of geometric guidance. First, we provide the geometry of surfaces where the current reconstruction differs from the complete model as a point cloud. This helps the framework to focus on regions that still need work. Second, we use geometric analysis to extract a set of planar prompts, that correspond to candidate surfaces where a CAD extrusion step could be started. Third, we present a step-wise sampling to generate multiple complete candidate CAD modeling steps instead of single-tokens without direct geometric interpretation. Our framework has three major components. Geometric guidance computation extracts the first two types of geometric guidance. Single-step reconstruction computes a single candidate CAD modeling step for each provided prompt. Single-step selection selects among the candidate CAD modeling steps. The process continues until the reconstruction is completed. Our quantitative results show a significant improvement across all metrics. For example, on the dataset DeepCAD, PS-CAD improves upon the best published SOTA method by reducing the geometry errors (CD and HD) by 10%, and the structural error (ECD metric) by about 13%.

Förderung

  • National Natural Science Foundation of China
  • National Key R&D Program of China
  • Beijing Nova Program, Beijing Natural Science Foundation
  • China Postdoctoral Science Foundation

Zitieren

APA

Yang, B., Jiang, H., Pan, H., Lin, G., Xiao, J., & Wonka, P. (2025). PS-CAD: Local Geometry Guidance via Prompting and Selection for CAD Reconstruction. ACM Transactions on Graphics, 44(3), 1-21. https://doi.org/10.1145/3733595

Vancouver

Yang B, Jiang H, Pan H, Lin G, Xiao J, Wonka P. PS-CAD: Local Geometry Guidance via Prompting and Selection for CAD Reconstruction. ACM Transactions on Graphics. 2025;44(3):1-21. doi:10.1145/3733595

BibTeX
@article{Yang2025,
  author = {Yang, Bingchen and Jiang, Haiyong and Pan, Hao and Lin, Guosheng and Xiao, Jun and Wonka, Peter},
  title = {PS-CAD: Local Geometry Guidance via Prompting and Selection for CAD Reconstruction},
  journal = {ACM Transactions on Graphics},
  year = {2025},
  volume = {44},
  number = {3},
  pages = {1--21},
  doi = {10.1145/3733595}
}
Kostenlos starten →