IGF-Fit: Implicit gradient field fitting for point cloud normal estimation
We introduce IGF-Fit, a novel method for estimating surface normals from point clouds with varying noise and density. Unlike previous approaches that rely on point-wise weights and explicit representations, IGF-Fit employs a network that learns an implicit representation and uses derivatives to pred...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2024-06-01
|
Series: | Graphical Models |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S152407032400002X |
_version_ | 1797289740855148544 |
---|---|
author | Bowen Lyu Li-Yong Shen Chun-Ming Yuan |
author_facet | Bowen Lyu Li-Yong Shen Chun-Ming Yuan |
author_sort | Bowen Lyu |
collection | DOAJ |
description | We introduce IGF-Fit, a novel method for estimating surface normals from point clouds with varying noise and density. Unlike previous approaches that rely on point-wise weights and explicit representations, IGF-Fit employs a network that learns an implicit representation and uses derivatives to predict normals. The input patch serves as both a shape latent vector and query points for fitting the implicit representation. To handle noisy input, we introduce a novel noise transformation module with a training strategy for noise classification and latent vector bias prediction. Our experiments on synthetic and real-world scan datasets demonstrate the effectiveness of IGF-Fit, achieving state-of-the-art performance on both noise-free and density-varying data. |
first_indexed | 2024-03-07T19:10:24Z |
format | Article |
id | doaj.art-6d65232884e44f8cb02f4301a265c8dc |
institution | Directory Open Access Journal |
issn | 1524-0703 |
language | English |
last_indexed | 2024-03-07T19:10:24Z |
publishDate | 2024-06-01 |
publisher | Elsevier |
record_format | Article |
series | Graphical Models |
spelling | doaj.art-6d65232884e44f8cb02f4301a265c8dc2024-03-01T05:05:46ZengElsevierGraphical Models1524-07032024-06-01133101214IGF-Fit: Implicit gradient field fitting for point cloud normal estimationBowen Lyu0Li-Yong Shen1Chun-Ming Yuan2School of Mathematical Sciences, University of Chinese Academy of Sciences, 100049, Beijing, ChinaSchool of Mathematical Sciences, University of Chinese Academy of Sciences, 100049, Beijing, China; Corresponding author.School of Mathematical Sciences, University of Chinese Academy of Sciences, 100049, Beijing, China; KLMM, Academy of Mathematics and Systems Sciences, CAS, 100190, Beijing, ChinaWe introduce IGF-Fit, a novel method for estimating surface normals from point clouds with varying noise and density. Unlike previous approaches that rely on point-wise weights and explicit representations, IGF-Fit employs a network that learns an implicit representation and uses derivatives to predict normals. The input patch serves as both a shape latent vector and query points for fitting the implicit representation. To handle noisy input, we introduce a novel noise transformation module with a training strategy for noise classification and latent vector bias prediction. Our experiments on synthetic and real-world scan datasets demonstrate the effectiveness of IGF-Fit, achieving state-of-the-art performance on both noise-free and density-varying data.http://www.sciencedirect.com/science/article/pii/S152407032400002XPoint cloudNormal estimationDeep learningImplicit representation |
spellingShingle | Bowen Lyu Li-Yong Shen Chun-Ming Yuan IGF-Fit: Implicit gradient field fitting for point cloud normal estimation Graphical Models Point cloud Normal estimation Deep learning Implicit representation |
title | IGF-Fit: Implicit gradient field fitting for point cloud normal estimation |
title_full | IGF-Fit: Implicit gradient field fitting for point cloud normal estimation |
title_fullStr | IGF-Fit: Implicit gradient field fitting for point cloud normal estimation |
title_full_unstemmed | IGF-Fit: Implicit gradient field fitting for point cloud normal estimation |
title_short | IGF-Fit: Implicit gradient field fitting for point cloud normal estimation |
title_sort | igf fit implicit gradient field fitting for point cloud normal estimation |
topic | Point cloud Normal estimation Deep learning Implicit representation |
url | http://www.sciencedirect.com/science/article/pii/S152407032400002X |
work_keys_str_mv | AT bowenlyu igffitimplicitgradientfieldfittingforpointcloudnormalestimation AT liyongshen igffitimplicitgradientfieldfittingforpointcloudnormalestimation AT chunmingyuan igffitimplicitgradientfieldfittingforpointcloudnormalestimation |