D$^2$CSG: Unsupervised Learning of Compact CSG Trees with Dual Complements and Dropouts
This addresses the challenge of generating compact and accurate CSG trees for 3D CAD modeling, which is incremental over prior neural CSG models.
The paper tackles the problem of unsupervised learning of compact constructive solid geometry (CSG) representations for 3D CAD shapes, achieving superior quality and more natural primitives compared to existing alternatives, especially for complex and high-genus shapes.
We present D$^2$CSG, a neural model composed of two dual and complementary network branches, with dropouts, for unsupervised learning of compact constructive solid geometry (CSG) representations of 3D CAD shapes. Our network is trained to reconstruct a 3D shape by a fixed-order assembly of quadric primitives, with both branches producing a union of primitive intersections or inverses. A key difference between D$^2$CSG and all prior neural CSG models is its dedicated residual branch to assemble the potentially complex shape complement, which is subtracted from an overall shape modeled by the cover branch. With the shape complements, our network is provably general, while the weight dropout further improves compactness of the CSG tree by removing redundant primitives. We demonstrate both quantitatively and qualitatively that D$^2$CSG produces compact CSG reconstructions with superior quality and more natural primitives than all existing alternatives, especially over complex and high-genus CAD shapes.