arrow
返回

PCGen: A Fully Parallelizable Point Cloud Generative Model

delete2024-02-22
delete0
delete
OA
AI
N
Nicolas Vercheval *
R
Remco Royen
A
Adrian Munteanu
A
Aleksandra Pižurica
DOI:10.3390/s24051414delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
Generative models have the potential to revolutionize 3D extended reality. A primary obstacle is that augmented and virtual reality need real-time computing. Current state-of-the-art point cloud random generation methods are not fast enough for these applications. We introduce a vector-quantized variational autoencoder model (VQVAE) that can synthesize high-quality point clouds in milliseconds. Unlike previous work in VQVAEs, our model offers a compact sample representation suitable for conditional generation and data exploration with potential applications in rapid prototyping. We achieve this result by combining architectural improvements with an innovative approach for probabilistic random generation. First, we rethink current parallel point cloud autoencoder structures, and we propose several solutions to improve robustness, efficiency and reconstruction quality. Notable contributions in the decoder architecture include an innovative computation layer to process the shape semantic information, an attention mechanism that helps the model focus on different areas and a filter to cover possible sampling errors. Secondly, we introduce a parallel sampling strategy for VQVAE models consisting of a double encoding system, where a variational autoencoder learns how to generate the complex discrete distribution of the VQVAE, not only allowing quick inference but also describing the shape with a few global variables. We compare the proposed decoder and our VQVAE model with established and concurrent work, and we prove, one by one, the validity of the single contributions.
Keyword:
point clouds
autoencoder
variational autoencoder
vector-quantized variational autoencoder
real-time computing
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Sensors 封面图
Sensors
IF:
3.5
论文数:
7.3W
被引数:
26.5W

机构

G
Ghent University
学者数:
2.8K
论文数: 928
被引数: 3.4W
V
Vrije Universiteit Brussel
学者数:
1.1K
论文数: 305
被引数: 8.2K
引用论文

引用论文

err分享
err收藏
Deep Unsupervised Learning of 3D Point Clouds via Graph Topology Inference and Filtering
err2020-01-01
err52
errOAAI
errChen, Siheng; Duan, Chaojing; Yang, Yaoqing; Li, Duanshun; Feng, Chen; Tian, Dong
err分享
err收藏
err分享
err收藏
IP Header Compression over PPP
err
IF0
err2003-07-01
err0
PREAI
errT. Koren; S. Casner; C. Bormann
err分享
err收藏
err
IF0
err
err0
PREAI
err
err分享
err收藏
err分享
err收藏
学者 查看更多内容