Point2CAD: Reverse Engineering CAD Models from 3D Point Clouds
Yujia Liu, Anton Obukhov, Jan Dirk Wegner, Konrad Schindler
Abstract
Figure 1 . Point2CAD is a pipeline for reconstructing complex CAD models, including freeform surfaces, from 3D point clouds. A raw point cloud is segmented into clusters corresponding to CAD faces, and each face is fitted either with a geometric primitive or with a parametric freeform surface, using a novel, implicit neural representation. Due to the analytic representation the surfaces can be extended and intersected, such that topology emerges, which is then used to clip the surface primitives.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext 275dfaa8-39ff-4a0c-9db3-db7f9fa83870Cited by top-tier papers17
- Split-and-Fit: Learning B-Reps via Structure-Aware Voronoi PartitioningYilin Liu, Jiale Chen, Shanshan Pan, Daniel Cohen-Or et al.SIGGRAPH 2024 · 23 citations
- SpelsNet: Surface Primitive Elements Segmentation by B-Rep Graph Structure SupervisionKseniya Cherenkova, Elona Dupont, Anis Kacem, Gleb Gusev et al.NeurIPS 2024 · 8 citations
- MV2Cyl: Reconstructing 3D Extrusion Cylinders from Multi-View ImagesEunji Hong, Minh Hieu Nguyen, Mikaela Angelina Uy, Minhyuk SungNeurIPS 2024 · 5 citations
- Drawing2CAD: Sequence-to-Sequence Learning for CAD Generation from Vector DrawingsFeiwei Qin, Shichao Lu, Junhao Hou, Changmiao Wang et al.ACM MM 2025 · 4 citations
- CADFit: Precise Mesh-to-CAD Program Generation with Hybrid OptimizationGhadi Nehme, Eamon Whalen, Faez AhmedICML 2026 · 2 citations
Builds on10
- Implicit Neural Representations with Periodic Activation FunctionsVincent Sitzmann, Julien N. P. Martel, Alexander W. Bergman, David B. Lindell et al.NeurIPS 2020 · 4,008 citations
- DeepGCNs: Can GCNs Go As Deep As CNNs?Guohao Li, Matthias Müller, Ali K. Thabet, Bernard GhanemICCV 2019 · 1,586 citations
- ComplexGen: CAD reconstruction by B-rep chain complex generationHaoxiang Guo, Shilin Liu, Hao Pan, Yang Liu et al.SIGGRAPH 2022 · 106 citations
- HPNet: Deep Primitive Segmentation Using Hybrid RepresentationsSiming Yan, Zhenpei Yang, Chongyang Ma, Haibin Huang et al.ICCV 2021 · 67 citations
- Point2Cyl: Reverse Engineering 3D Objects from Point Clouds to Extrusion CylindersMikaela Angelina Uy, Yen-Yu Chang, Minhyuk Sung, Purvi Goel et al.CVPR 2022 · 54 citations
Related papers
- Learning CAD Modeling Sequences via Projection and Part AwarenessYang Liu, Daxuan Ren, Yijie Ding, Jianmin Zheng et al.NeurIPS 2025
- CAPRI-Net: Learning Compact CAD Shapes with Adaptive Primitive AssemblyFenggen Yu, Zhiqin Chen, Manyi Li, Aditya Sanghi et al.CVPR 2022 · 53 citations
- Autoregressive B-Rep Shape Generation with Parametric SurfacesDafei Qin, Rui Xu, Zeyu Shen, Kaichun Qiao et al.SIGGRAPH 2026 · 2 citations
- SECAD-Net: Self-Supervised CAD Reconstruction by Learning Sketch-Extrude OperationsPu Li, Jianwei Guo, Xiaopeng Zhang, Dong-Ming YanCVPR 2023
- SfmCAD: Unsupervised CAD Reconstruction by Learning Sketch-based Feature Modeling OperationsPu Li, Jianwei Guo, Huibin Li, Bedrich Benes et al.CVPR 2024
