Shape Generation using Spatially Partitioned Point Clouds

TL;DR

This paper introduces a scalable point cloud-based method for 3D shape generation that combines spatial partitioning, PCA, and neural networks to produce high-quality, attribute-rich 3D models more efficiently than voxel-based approaches.

Contribution

The paper presents a novel point cloud-based shape generation framework that integrates spatial partitioning, PCA, and neural networks for improved efficiency and attribute incorporation.

Findings

Outperforms voxel-based models in quality and scalability.

Uses neural networks to model complex shape coefficient distributions.

Handles additional point attributes like normals and color.

Abstract

We propose a method to generate 3D shapes using point clouds. Given a point-cloud representation of a 3D shape, our method builds a kd-tree to spatially partition the points. This orders them consistently across all shapes, resulting in reasonably good correspondences across all shapes. We then use PCA analysis to derive a linear shape basis across the spatially partitioned points, and optimize the point ordering by iteratively minimizing the PCA reconstruction error. Even with the spatial sorting, the point clouds are inherently noisy and the resulting distribution over the shape coefficients can be highly multi-modal. We propose to use the expressive power of neural networks to learn a distribution over the shape coefficients in a generative-adversarial framework. Compared to 3D shape generative models trained on voxel-representations, our point-based method is considerably more light-weight and scalable, with little loss of quality. It also outperforms simpler linear factor models such as Probabilistic PCA, both qualitatively and quantitatively, on a number of categories from the ShapeNet dataset. Furthermore, our method can easily incorporate other point attributes such as normal and color information, an additional advantage over voxel-based representations.