A topology-based descriptor for 3D point cloud modeling: Theory and experiments

William J. Beksi, Nikolaos P Papanikolopoulos

Research output: Contribution to journalArticle

Abstract

This paper presents a topology-based global descriptor that allows for efficient 3D point cloud processing tasks associated with the analysis of shapes. The descriptor is called the Signature of Topologically Persistent Points (STPP). By using persistent homology, STPP is formed by the computation of topological invariants involving the zeroth and first homology groups. Persistent homology is a methodology that finds the features of a topological space at different spatial resolutions. STPP requires no preprocessing and uses a single tuning parameter. It is an effective 3D point cloud descriptor with robustness to noisy sensor data. The paper highlights this aspect with experimental comparisons to the state of the art. Our research has been validated on a publicly available RGB-D dataset. The results show that STPP can be used as a distinctive signature by employing a small number of features with object detection and classification benefiting from its usage.

Original languageEnglish (US)
Pages (from-to)84-95
Number of pages12
JournalImage and Vision Computing
Volume88
DOIs
StatePublished - Aug 1 2019

Fingerprint

Tuning
Topology
Sensors
Processing
Experiments
Object detection

Keywords

  • Object classification
  • Persistent homology
  • Shape analysis
  • Topological data analysis

Cite this

A topology-based descriptor for 3D point cloud modeling : Theory and experiments. / Beksi, William J.; Papanikolopoulos, Nikolaos P.

In: Image and Vision Computing, Vol. 88, 01.08.2019, p. 84-95.

Research output: Contribution to journalArticle

@article{6947b3f96c004bf3a655d877b1d32924,
title = "A topology-based descriptor for 3D point cloud modeling: Theory and experiments",
abstract = "This paper presents a topology-based global descriptor that allows for efficient 3D point cloud processing tasks associated with the analysis of shapes. The descriptor is called the Signature of Topologically Persistent Points (STPP). By using persistent homology, STPP is formed by the computation of topological invariants involving the zeroth and first homology groups. Persistent homology is a methodology that finds the features of a topological space at different spatial resolutions. STPP requires no preprocessing and uses a single tuning parameter. It is an effective 3D point cloud descriptor with robustness to noisy sensor data. The paper highlights this aspect with experimental comparisons to the state of the art. Our research has been validated on a publicly available RGB-D dataset. The results show that STPP can be used as a distinctive signature by employing a small number of features with object detection and classification benefiting from its usage.",
keywords = "Object classification, Persistent homology, Shape analysis, Topological data analysis",
author = "Beksi, {William J.} and Papanikolopoulos, {Nikolaos P}",
year = "2019",
month = "8",
day = "1",
doi = "10.1016/j.imavis.2019.05.004",
language = "English (US)",
volume = "88",
pages = "84--95",
journal = "Image and Vision Computing",
issn = "0262-8856",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - A topology-based descriptor for 3D point cloud modeling

T2 - Theory and experiments

AU - Beksi, William J.

AU - Papanikolopoulos, Nikolaos P

PY - 2019/8/1

Y1 - 2019/8/1

N2 - This paper presents a topology-based global descriptor that allows for efficient 3D point cloud processing tasks associated with the analysis of shapes. The descriptor is called the Signature of Topologically Persistent Points (STPP). By using persistent homology, STPP is formed by the computation of topological invariants involving the zeroth and first homology groups. Persistent homology is a methodology that finds the features of a topological space at different spatial resolutions. STPP requires no preprocessing and uses a single tuning parameter. It is an effective 3D point cloud descriptor with robustness to noisy sensor data. The paper highlights this aspect with experimental comparisons to the state of the art. Our research has been validated on a publicly available RGB-D dataset. The results show that STPP can be used as a distinctive signature by employing a small number of features with object detection and classification benefiting from its usage.

AB - This paper presents a topology-based global descriptor that allows for efficient 3D point cloud processing tasks associated with the analysis of shapes. The descriptor is called the Signature of Topologically Persistent Points (STPP). By using persistent homology, STPP is formed by the computation of topological invariants involving the zeroth and first homology groups. Persistent homology is a methodology that finds the features of a topological space at different spatial resolutions. STPP requires no preprocessing and uses a single tuning parameter. It is an effective 3D point cloud descriptor with robustness to noisy sensor data. The paper highlights this aspect with experimental comparisons to the state of the art. Our research has been validated on a publicly available RGB-D dataset. The results show that STPP can be used as a distinctive signature by employing a small number of features with object detection and classification benefiting from its usage.

KW - Object classification

KW - Persistent homology

KW - Shape analysis

KW - Topological data analysis

UR - http://www.scopus.com/inward/record.url?scp=85067251866&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067251866&partnerID=8YFLogxK

U2 - 10.1016/j.imavis.2019.05.004

DO - 10.1016/j.imavis.2019.05.004

M3 - Article

AN - SCOPUS:85067251866

VL - 88

SP - 84

EP - 95

JO - Image and Vision Computing

JF - Image and Vision Computing

SN - 0262-8856

ER -