Conditional Invertible Flow for Point Cloud Generation
This addresses the problem of generating realistic 3D point clouds for applications in computer vision and graphics, presenting a novel method but appearing incremental in its approach.
The paper tackles 3D point cloud generation by using invertible flow-based models to treat point clouds as probability densities, with each cloud defined by a small embedding vector and parameter sharing among networks, and evaluates the model's generative capabilities qualitatively and quantitatively.
This paper focuses on a novel generative approach for 3D point clouds that makes use of invertible flow-based models. The main idea of the method is to treat a point cloud as a probability density in 3D space that is modeled using a cloud-specific neural network. To capture the similarity between point clouds we rely on parameter sharing among networks, with each cloud having only a small embedding vector that defines it. We use invertible flows networks to generate the individual point clouds, and to regularize the embedding vectors. We evaluate the generative capabilities of the model both in qualitative and quantitative manner.