Skip to main content

Point cloud synthesis with stochastic differential equations.

Li, T., Wang, M., Liu, X., Liang, H., Chang, J. and Zhang, J. J., 2023. Point cloud synthesis with stochastic differential equations. Computer Animation and Virtual Worlds, 34 (5), e2140.

Full text available as:

Style_guidelines_for_CASA_Submissions_Proceedings_Paper_1_-3.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial.


DOI: 10.1002/cav.2140


In this article, we propose a point cloud synthesis method based on stochastic differential equations. We view the point cloud generation process as smoothly transforming from a known prior distribution toward the high-likelihood shape by point-level denoising. We introduce a conditional corrector sampler to improve the quality of point clouds. By leveraging Markov chain Monte Carlo sample, our method can synthesize realistic point clouds. We additionally prove that our approach can be trained in an auto-encoding fashion and reconstruct the point cloud faithfully. Furthermore, our model can be extended on a downstream application of point cloud completion. Experimental results demonstrate the effectiveness and efficiency of our method.

Item Type:Article
Uncontrolled Keywords:point cloud reconstruction; point cloud synthesis; stochastic differential equations
Group:Faculty of Media & Communication
ID Code:39355
Deposited By: Symplectic RT2
Deposited On:11 Jan 2024 13:09
Last Modified:26 Feb 2024 01:08


Downloads per month over past year

More statistics for this item...
Repository Staff Only -