Weakly supervised part segmentation of 3D models based on sphere node graph

Liuhong Lv; Yujie Lu; Yulong Wang; Zijian Wang; Shen Cai

doi:10.1117/12.3004174

11 October 2023 Weakly supervised part segmentation of 3D models based on sphere node graph

Liuhong Lv, Yujie Lu, Yulong Wang, Zijian Wang, Shen Cai

Proceedings Volume 12800, Sixth International Conference on Computer Information Science and Application Technology (CISAT 2023); 128004S (2023) https://doi.org/10.1117/12.3004174
Event: 6th International Conference on Computer Information Science and Application Technology (CISAT 2023), 2023, Hangzhou, China

Abstract

In this paper, a weakly supervised 3D model part segmentation method based on Sphere Node Graph (SN-Graph) is proposed. For a given 3D model (usually in the form of mesh), we extract the sphere node graph from its interior space as the discrimination criterion of surface points segmentation, since SN-Graph explicitly represents the global and local structures. These sphere nodes are unsupervised, while the weakly supervised information for part segmentation comes from a small number of labelled point clouds for each model in the training set. Utilizing these small number of labelled points, we train the bisection segmentation information for each sphere by parameterizing the orientation of the plane through the center of the sphere and classification features. Experiments demonstrate that the proposed method significantly improves the segmentation accuracy of surface points at part junctions and achieves similar overall accuracy to fully supervised methods (utilizing more than 1K labelled points).

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Liuhong Lv, Yujie Lu, Yulong Wang, Zijian Wang, and Shen Cai "Weakly supervised part segmentation of 3D models based on sphere node graph", Proc. SPIE 12800, Sixth International Conference on Computer Information Science and Application Technology (CISAT 2023), 128004S (11 October 2023); https://doi.org/10.1117/12.3004174

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Optical spheres

3D modeling

Point clouds

Education and training

Image segmentation

Network architectures

Visualization

Show All Keywords

Keywords/Phrases

Search In:

Publication Years