Point2Cyl: Reverse Engineering 3D Objects from Point Clouds to Extrusion Cylinders

Uy, Mikaela Angelina; Chang, Yen-yu; Sung, Minhyuk; Goel, Purvi; Lambourne, Joseph G.; Birdal, Tolga; Guibas, Leonidas J.

doi:10.48550/arxiv.2112.09329

Cited by 1 publication

(1 citation statement)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another line of research leverages the supervision available from modeling operations stored in parametric CAD files to directly predict sequences of CAD modeling operations. The result is editable CAD files in the form of 2D engineering sketches [11,26,29,42] or 3D CAD models generated from a B-rep [43,46] or point cloud [36] target, interactive system [19], or generative model [44]. Although sequential CAD generation approaches have made significant progress, an outstanding challenge is to extend these techniques to other modeling operations beyond sketch and extrude.…”

Section: Related Workmentioning

confidence: 99%

SolidGen: An Autoregressive Model for Direct B-rep Synthesis

Jayaraman¹,

Lambourne²,

Desai³

et al. 2022

Preprint

View full text Add to dashboard Cite

The Boundary representation (B-rep) format is the de-facto shape representation in computer-aided design (CAD) to model watertight solid objects. Recent approaches to generating CAD models have focused on learning sketch-and-extrude modeling sequences that are executed by a solid modeling kernel in postprocess to recover a B-rep. In this paper we present a new approach that enables learning from and synthesizing B-reps without the need for supervision through CAD modeling sequence data. Our method SolidGen, is an autoregressive neural network that models the B-rep directly by predicting the vertices, edges and faces using Transformer-based and pointer neural networks. Key to achieving this is our Indexed Boundary Representation that references B-rep vertices, edges and faces in a well-defined hierarchy to capture the geometric and topological relations suitable for use with machine learning. SolidGen can be easily conditioned on contexts e.g., class labels thanks to its probabilistic modeling of the B-rep distribution. We demonstrate qualitatively, quantitatively and through perceptual evaluation by human subjects that SolidGen can produce high quality, realistic looking CAD models.

show abstract