APSIPA Transactions on Signal and Information Processing > Vol 14 > Issue 2

LPSR: Lightweight Point Cloud Surface Reconstruction

Qingyang Zhou, University of Southern California, USA, qzhou776@usc.edu , Chee-An Yu, University of Southern California, USA, Xuechun Hua, University of Southern California, USA, Shan Liu, Tencent Media Lab, USA, C.-C. Jay Kuo, University of Southern California, USA
 
Suggested Citation
Qingyang Zhou, Chee-An Yu, Xuechun Hua, Shan Liu and C.-C. Jay Kuo (2025), "LPSR: Lightweight Point Cloud Surface Reconstruction", APSIPA Transactions on Signal and Information Processing: Vol. 14: No. 2, e105. http://dx.doi.org/10.1561/116.20240069

Publication Date: 23 Apr 2025
© 2025 Q. Zhou, C.-A. Yu, X. Hua, S. Liu and C.-C. J. Kuo
 
Subjects
3D reconstruction and image-based modeling,  Shape,  Shape representation
 
Keywords
Surface reconstructionpoint cloudsshape modelinggreen learning
 

Share

Open Access

This is published under the terms of CC BY-NC.

Downloaded: 46 times

In this article:
Introduction 
Related Work 
Proposed LPSR Method 
Experiments 
Conclusion and Future Work 
Acknowledgments 
References 

Abstract

Surface reconstruction from point cloud scans is crucial in 3D vision and graphics. Recent approaches focus on training deep-learning (DL) models to generate representations through learned priors. These models use neural networks to map point clouds into compact representations and then decode these latent representations into signed distance functions (SDFs). Such methods rely on heavy supervision and incur high computational costs. Moreover, they lack interpretability regarding how the encoded representations influence the resulting surfaces. This work proposes a computationally efficient and mathematically transparent Green Learning (GL) solution. We name it the lightweight pointcloud surface reconstruction (LPSR) method. LPSR reconstructs surfaces in two steps. First, it progressively generates a sparse voxel representation using a feedforward approach. Second, it decodes the representation into unsigned distance functions (UDFs) based on anisotropic heat diffusion. Experimental results show that LPSR offers competitive performance against state-of-theart surface reconstruction methods on the FAMOUS, ABC, and Thingi10K datasets at modest model complexity.

DOI:10.1561/116.20240069

Companion

APSIPA Transactions on Signal and Information Processing Special Issue - Three-dimensional Point Cloud Data Modeling, Processing, and Analysis
See the other articles that are part of this special issue.