Geometry-Informed Neural Networks
This addresses the challenge of data scarcity in geometry tasks for fields like computer graphics and engineering design, offering a novel generative approach that is incremental in leveraging existing neural field methods.
The paper tackles the problem of limited shape datasets for supervised learning by introducing geometry-informed neural networks (GINNs), a framework that trains shape-generative models without data using user-specified objectives and constraints, and it shows the ability to generate multiple diverse solutions with control over properties like surface smoothness and hole count.
Geometry is a ubiquitous tool in computer graphics, design, and engineering. However, the lack of large shape datasets limits the application of state-of-the-art supervised learning methods and motivates the exploration of alternative learning strategies. To this end, we introduce geometry-informed neural networks (GINNs) -- a framework for training shape-generative neural fields without data by leveraging user-specified design requirements in the form of objectives and constraints. By adding diversity as an explicit constraint, GINNs avoid mode-collapse and can generate multiple diverse solutions, often required in geometry tasks. Experimentally, we apply GINNs to several problems spanning physics, geometry, and engineering design, showing control over geometrical and topological properties, such as surface smoothness or the number of holes. These results demonstrate the potential of training shape-generative models without data, paving the way for new generative design approaches without large datasets.