Geometry and Learning Co-Supported Normal Estimation for Unstructured Point Cloud

Haoran ZHOU, Honghua CHEN, Yidan FENG, Qiong WANG, Jing QIN, Haoran XIE, Fu Lee WANG, Mingqiang WEI, Jun WANG

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)peer-review

Abstract

In this paper, we propose a normal estimation method for unstructured point cloud. We observe that geometric estimators commonly focus more on feature preservation but are hard to tune parameters and sensitive to noise, while learning-based approaches pursue an overall normal estimation accuracy but cannot well handle challenging regions such as surface edges. This paper presents a novel normal estimation method, under the co-support of geometric estimator and deep learning. To lowering the learning difficulty, we first propose to compute a suboptimal initial normal at each point by searching for a best fitting patch. Based on the computed normal field, we design a normal-based height map network (NH-Net) to fine-tune the suboptimal normals. Qualitative and quantitative evaluations demonstrate the clear improvements of our results over both traditional methods and learning-based methods, in terms of estimation accuracy and feature recovery.
Original languageEnglish
Title of host publicationThe Proceedings of the Conference on Computer Vision and Pattern Recognition 2020 (CVPR 2020)
Pages13235-13244
Number of pages10
DOIs
Publication statusPublished - 14 Jun 2020
EventThe IEEE/CVF Conference on Computer Vision and Pattern Recognition 2020 - Online
Duration: 14 Jun 202019 Jun 2020
http://cvpr2020.thecvf.com/

Publication series

NameProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
PublisherIEEE Computer Society
ISSN (Print)1063-6919

Public Lecture

Public LectureThe IEEE/CVF Conference on Computer Vision and Pattern Recognition 2020
Abbreviated titleCVPR2020
Period14/06/2019/06/20
Internet address

Bibliographical note

Funding Information:
This work was supported by the National Natural Science Foundation of China (No. 61502137), the Hong Kong Research Grants Council (No. PolyU 152035/17E), the HKIBS Research Seed Fund 2019/20 (No. 190-009), and the Research Seed Fund (No. 102367) of Lingnan University, Hong Kong.

Fingerprint Dive into the research topics of 'Geometry and Learning Co-Supported Normal Estimation for Unstructured Point Cloud'. Together they form a unique fingerprint.

Cite this