CvxNet: Learnable Convex Decomposition
Any solid object can be decomposed into a collection of convex polytopes (in short, convexes). When a small number of convexes are used, such a decomposition can bethought of as a piece-wise approximation of the geometry.This decomposition is fundamental to real-time physics simulation in computer graphics, where it creates a unified representation of dynamic geometry for collision detection. A convex object also has the property of being simultaneously an explicit and implicit representation: one can interpret it explicitly as a mesh derived by computing the vertices of a convex hull, or implicitly as the collection of half-space constraints or support functions. Their implicit representation makes them particularly well suited for neural net-work training, as they abstract away from the topology of the geometry they need to represent. We introduce a net-work architecture to represent a low dimensional family of convexes. This family is automatically derived via an auto-encoding process. We investigate the applications of this architecture including automatic convex decomposition, image to 3D reconstruction, and part-based shape retrieval.
READ FULL TEXT