Partial Symmetry Detection in Large-Scale 3D Data Collections
Abstract/Contents
- Abstract
- This thesis describes a component-level discrete symmetry detection algorithm that detects translation, reflection, and rotation symmetry in pairs of parts of a whole shape. The dataset used is ShapeNetCore, a large-scale dataset of man-made 3D shapes of everyday objects. Currently, ShapeNet only provides information on pre-segmentation of the shapes. We are interested in improving ShapeNet by annotating the shapes with information on the symmetries within the models. In addition to the three partial symmetries listed before, the algorithm also detects global self-reflection as well as symmetry groups of pairs of parts. The results show the partial symmetry detection algorithm at around 99% accuracy for detecting reflection and rotation symmetries and around 95% for translation symmetries on the dataset of chair shapes. Future work includes evaluating the algorithm on other kinds of shapes, as well as improving the current partial symmetry detection to reduce errors. Future uses of the partial symmetry detection algorithm include helping with 3D shape completion from partial shapes.
Description
Type of resource | text |
---|---|
Date created | May 6, 2019 |
Creators/Contributors
Author | Yau, Jacqueline H. |
---|---|
Degree granting institution | Stanford University, Department of Computer Science |
Primary advisor | Guibas, Leonidas J. |
Advisor | Dubrovina, Anastasia |
Subjects
Subject | symmetry |
---|---|
Subject | ShapeNetCore |
Subject | computer science |
Subject | discrete symmetries |
Subject | graphics |
Subject | component-level |
Genre | Thesis |
Bibliographic information
Related item |
|
---|---|
Location | https://purl.stanford.edu/fm115yy3522 |
Access conditions
- Use and reproduction
- User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Preferred citation
- Preferred Citation
- Yau, Jacqueline H. (2019). Partial Symmetry Detection in Large-Scale 3D Data Collections. Stanford Digital Repository. Available at: https://purl.stanford.edu/fm115yy3522
Collection
Undergraduate Theses, School of Engineering
View other items in this collection in SearchWorksContact information
- Contact
- jyau@cs.stanford.edu
Also listed in
Loading usage metrics...