Covariance based point cloud descriptors for object detection and recognition

Duc Fehr, William J. Beksi, Dimitris Zermas, Nikolaos P Papanikolopoulos

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

Processing 3D point cloud data is of primary interest in many areas of computer vision, including object grasping, robot navigation, and object recognition. The introduction of affordable RGB-D sensors has created a great interest in the computer vision community towards developing efficient algorithms for point cloud processing. Previously, capturing a point cloud required expensive specialized sensors such as lasers or dedicated range imaging devices; now, range data is readily available from low-cost sensors that provide easily extractable point clouds from a depth map. From here, an interesting challenge is to find different objects in the point cloud. Various descriptors have been introduced to match features in a point cloud. Cheap sensors are not necessarily designed to produce precise measurements, which means that the data is not as accurate as a point cloud provided from a laser or a dedicated range finder. Although some feature descriptors have been shown to be successful in recognizing objects from point clouds, there still exists opportunities for improvement. The aim of this paper is to introduce techniques from other fields, such as image processing, into 3D point cloud processing in order to improve rendering, classification, and recognition. Covariances have proven to be a success not only in image processing, but in other domains as well. This work develops the application of covariances in conjunction with 3D point cloud data.

Original languageEnglish (US)
Pages (from-to)80-93
Number of pages14
JournalComputer Vision and Image Understanding
Volume142
DOIs
StatePublished - Jan 1 2016

Fingerprint

Object recognition
Sensors
Computer vision
Image processing
Processing
Range finders
Lasers
Navigation
Robots
Imaging techniques
Object detection
Costs

Keywords

  • Classification
  • Colored point clouds
  • Object recognition
  • RGB-D data

Cite this

Covariance based point cloud descriptors for object detection and recognition. / Fehr, Duc; Beksi, William J.; Zermas, Dimitris; Papanikolopoulos, Nikolaos P.

In: Computer Vision and Image Understanding, Vol. 142, 01.01.2016, p. 80-93.

Research output: Contribution to journalArticle

@article{efa8ec9af3fd4d8d8ef0a41853406cbf,
title = "Covariance based point cloud descriptors for object detection and recognition",
abstract = "Processing 3D point cloud data is of primary interest in many areas of computer vision, including object grasping, robot navigation, and object recognition. The introduction of affordable RGB-D sensors has created a great interest in the computer vision community towards developing efficient algorithms for point cloud processing. Previously, capturing a point cloud required expensive specialized sensors such as lasers or dedicated range imaging devices; now, range data is readily available from low-cost sensors that provide easily extractable point clouds from a depth map. From here, an interesting challenge is to find different objects in the point cloud. Various descriptors have been introduced to match features in a point cloud. Cheap sensors are not necessarily designed to produce precise measurements, which means that the data is not as accurate as a point cloud provided from a laser or a dedicated range finder. Although some feature descriptors have been shown to be successful in recognizing objects from point clouds, there still exists opportunities for improvement. The aim of this paper is to introduce techniques from other fields, such as image processing, into 3D point cloud processing in order to improve rendering, classification, and recognition. Covariances have proven to be a success not only in image processing, but in other domains as well. This work develops the application of covariances in conjunction with 3D point cloud data.",
keywords = "Classification, Colored point clouds, Object recognition, RGB-D data",
author = "Duc Fehr and Beksi, {William J.} and Dimitris Zermas and Papanikolopoulos, {Nikolaos P}",
year = "2016",
month = "1",
day = "1",
doi = "10.1016/j.cviu.2015.06.008",
language = "English (US)",
volume = "142",
pages = "80--93",
journal = "Computer Vision and Image Understanding",
issn = "1077-3142",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Covariance based point cloud descriptors for object detection and recognition

AU - Fehr, Duc

AU - Beksi, William J.

AU - Zermas, Dimitris

AU - Papanikolopoulos, Nikolaos P

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Processing 3D point cloud data is of primary interest in many areas of computer vision, including object grasping, robot navigation, and object recognition. The introduction of affordable RGB-D sensors has created a great interest in the computer vision community towards developing efficient algorithms for point cloud processing. Previously, capturing a point cloud required expensive specialized sensors such as lasers or dedicated range imaging devices; now, range data is readily available from low-cost sensors that provide easily extractable point clouds from a depth map. From here, an interesting challenge is to find different objects in the point cloud. Various descriptors have been introduced to match features in a point cloud. Cheap sensors are not necessarily designed to produce precise measurements, which means that the data is not as accurate as a point cloud provided from a laser or a dedicated range finder. Although some feature descriptors have been shown to be successful in recognizing objects from point clouds, there still exists opportunities for improvement. The aim of this paper is to introduce techniques from other fields, such as image processing, into 3D point cloud processing in order to improve rendering, classification, and recognition. Covariances have proven to be a success not only in image processing, but in other domains as well. This work develops the application of covariances in conjunction with 3D point cloud data.

AB - Processing 3D point cloud data is of primary interest in many areas of computer vision, including object grasping, robot navigation, and object recognition. The introduction of affordable RGB-D sensors has created a great interest in the computer vision community towards developing efficient algorithms for point cloud processing. Previously, capturing a point cloud required expensive specialized sensors such as lasers or dedicated range imaging devices; now, range data is readily available from low-cost sensors that provide easily extractable point clouds from a depth map. From here, an interesting challenge is to find different objects in the point cloud. Various descriptors have been introduced to match features in a point cloud. Cheap sensors are not necessarily designed to produce precise measurements, which means that the data is not as accurate as a point cloud provided from a laser or a dedicated range finder. Although some feature descriptors have been shown to be successful in recognizing objects from point clouds, there still exists opportunities for improvement. The aim of this paper is to introduce techniques from other fields, such as image processing, into 3D point cloud processing in order to improve rendering, classification, and recognition. Covariances have proven to be a success not only in image processing, but in other domains as well. This work develops the application of covariances in conjunction with 3D point cloud data.

KW - Classification

KW - Colored point clouds

KW - Object recognition

KW - RGB-D data

UR - http://www.scopus.com/inward/record.url?scp=84947040694&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947040694&partnerID=8YFLogxK

U2 - 10.1016/j.cviu.2015.06.008

DO - 10.1016/j.cviu.2015.06.008

M3 - Article

VL - 142

SP - 80

EP - 93

JO - Computer Vision and Image Understanding

JF - Computer Vision and Image Understanding

SN - 1077-3142

ER -