WPDv2: Implicit‐Assisted Poisson‐Disk Resampling with Sharp Feature Preservation

Abstract Point cloud resampling aims to reduce redundancy while preserving geometric structures. However, practical inputs often pose challenges such as noise, incompleteness, and non‐uniform distributions. These degraded conditions expose the limitations of existing methods, which cannot simultaneously ensure global coverage, regularity and feature preservation. In this paper, we propose WPDv2, a training‐free and implicit‐assisted Poisson‐disk resampling framework that integrates implicit surface reconstruction with sharpness‐aware feature weighting. It leverages implicit reconstruction to establish a coherent geometric domain and employs a winding number‐guided sharpness field to drive density‐adaptive sampling. Additionally, a feature‐constrained Voronoi relaxation enhances spatial regularity while preserving distinct geometric structures with sharp features. Extensive experiments demonstrate that WPDv2 achieves superior feature preservation, robust performance under noisy and incomplete inputs. It provides a practical solution for high‐fidelity point cloud resampling tasks.

Authors

Institutions

Publication Details

Journal
Computer Graphics Forum
Published
2026-11-13
DOI
https://doi.org/10.1111/cgf.70627
Primary Topic
3D Shape Modeling and Analysis
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

WPDv2: Implicit‐Assisted Poisson‐Disk Resampling with Sharp Feature Preservation

Baodong Wang, Yong‐Jin Liu, Junli Zhao, Chenlei Lv et al.
2 links
Computer Graphics Forum
3D Shape Modeling and Analysis
article

WPDv2: Implicit‐Assisted Poisson‐Disk Resampling with Sharp Feature Preservation

Baodong Wang, Yong‐Jin Liu, Junli Zhao, Chenlei Lv, Zhongke Wu, Xiaotan Sun
article en

Abstract

Abstract Point cloud resampling aims to reduce redundancy while preserving geometric structures. However, practical inputs often pose challenges such as noise, incompleteness, and non‐uniform distributions. These degraded conditions expose the limitations of existing methods, which cannot simultaneously ensure global coverage, regularity and feature preservation. In this paper, we propose WPDv2, a training‐free and implicit‐assisted Poisson‐disk resampling framework that integrates implicit surface reconstruction with sharpness‐aware feature weighting. It leverages implicit reconstruction to establish a coherent geometric domain and employs a winding number‐guided sharpness field to drive density‐adaptive sampling. Additionally, a feature‐constrained Voronoi relaxation enhances spatial regularity while preserving distinct geometric structures with sharp features. Extensive experiments demonstrate that WPDv2 achieves superior feature preservation, robust performance under noisy and incomplete inputs. It provides a practical solution for high‐fidelity point cloud resampling tasks.

Computer Graphics Forum
Qingdao University (CN), Shenzhen University (CN), Beijing Normal University (CN), Tsinghua University (CN)
National Natural Science Foundation of China, Natural Science Foundation of Shandong Province, Shenzhen Fundamental Research Program, Basic and Applied Basic Research Foundation of Guangdong Province
3D Shape Modeling and Analysis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon