3D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutions

Dong Wook Shu, Sung Woo Park, Junseok Kwon; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 3859-3868

Abstract


In this paper, we propose a novel generative adversarial network (GAN) for 3D point clouds generation, which is called tree-GAN. To achieve state-of-the-art performance for multi-class 3D point cloud generation, a tree-structured graph convolution network (TreeGCN) is introduced as a generator for tree-GAN. Because TreeGCN performs graph convolutions within a tree, it can use ancestor information to boost the representation power for features. To evaluate GANs for 3D point clouds accurately, we develop a novel evaluation metric called Frechet point cloud distance (FPD). Experimental results demonstrate that the proposed tree-GAN outperforms state-of-the-art GANs in terms of both conventional metrics and FPD, and can generate point clouds for different semantic parts without prior knowledge.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Shu_2019_ICCV,
author = {Shu, Dong Wook and Park, Sung Woo and Kwon, Junseok},
title = {3D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutions},
booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
month = {October},
year = {2019}
}