Animating high-fidelity video portrait with speech audio is crucial for virtual reality and digital
entertainment. While most previous studies rely on accurate explicit structural information …

We view the reconstruction of CAD models in the boundary representation (B-Rep) as the
detection of geometric primitives of different orders, ie, vertices, edges and surface patches …

Abstract We propose Point2Cyl, a supervised network transforming a raw 3D point cloud to a
set of extrusion cylinders. Reverse engineering from a raw geometry to a CAD model is an …

This paper presents BrepGen, a diffusion-based generative approach that directly outputs a
Boundary representation (B-rep) Computer-Aided Design (CAD) model. BrepGen …

Reverse engineering CAD models from raw geometry is a classic but strenuous research
problem. Previous learning-based methods rely heavily on labels due to the supervised …

We introduce CAPRI-Net, a self-supervised neural network for learning compact and
interpretable implicit representations of 3D computer-aided design (CAD) models, in the …

Sketch-and-extrude is a common and intuitive modeling process in computer aided design.
This paper studies the problem of learning the shape given in the form of point clouds by …

Programs offer compactness and structure that makes them an attractive representation for
visual data. We explore how code rewriting can be used to improve systems for inferring …

We introduce ShapeCoder, the first system capable of taking a dataset of shapes,
represented with unstructured primitives, and jointly discovering (i) useful abstraction …

Procedural models (ie symbolic programs that output visual data) are a historically‐popular
method for representing graphics content: vegetation, buildings, textures, etc. They offer …