A Progressive Conditional Generative Adversarial Network for Generating Dense and Colored 3D Point Clouds

TL;DR

This paper presents a progressive conditional GAN that generates dense, colored 3D point clouds with fine details across multiple resolutions, using a novel point transformer and graph convolutions.

Contribution

It introduces a progressive training method with a point transformer for high-resolution 3D point cloud generation, advancing the quality and detail of generated data.

Findings

Capable of learning and mimicking 3D data distribution.

Produces detailed colored point clouds at multiple resolutions.

Uses a novel progressive growth strategy with graph convolutions.

Abstract

In this paper, we introduce a novel conditional generative adversarial network that creates dense 3D point clouds, with color, for assorted classes of objects in an unsupervised manner. To overcome the difficulty of capturing intricate details at high resolutions, we propose a point transformer that progressively grows the network through the use of graph convolutions. The network is composed of a leaf output layer and an initial set of branches. Every training iteration evolves a point vector into a point cloud of increasing resolution. After a fixed number of iterations, the number of branches is increased by replicating the last branch. Experimental results show that our network is capable of learning and mimicking a 3D data distribution, and produces colored point clouds with fine details at multiple resolutions.