Indoor scene reconstruction from sets of noisy range images

Jens Gregor, R. T. Whitaker

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

This paper describes a system for building 3D models of indoor scenes from sets of noisy laser range images. It addresses several important aspects of this problem, namely, preprocessing, which includes image segmentation and planar model fitting; view registration, which is the method of determining the rigid transformation that describes the relative pose of the camera platform between views; and reconstruction, which is the subsequent integration or fusion of separate range images into a single 3D model. Our proposed strategy is to use a statistical sensor model. We thus account for noise properties of the data at each stage in the reconstruction process, which produces reliable results even in presence of significant measurement noise. We given a empirical analysis of a plane-based registration method and present results using real range data that demonstrate the performance of the entire reconstruction system.

Original languageEnglish (US)
Pages (from-to)304-332
Number of pages29
JournalGraphical Models
Volume63
Issue number5
DOIs
StatePublished - Jan 1 2001

Fingerprint

Range Image
3D Model
Registration
Model Fitting
Empirical Analysis
Image Segmentation
Preprocessing
Fusion
Camera
Image segmentation
Entire
Laser
Sensor
Fusion reactions
Cameras
Range of data
Demonstrate
Lasers
Sensors
Model

All Science Journal Classification (ASJC) codes

  • Software
  • Modeling and Simulation
  • Geometry and Topology
  • Computer Graphics and Computer-Aided Design

Cite this

Indoor scene reconstruction from sets of noisy range images. / Gregor, Jens; Whitaker, R. T.

In: Graphical Models, Vol. 63, No. 5, 01.01.2001, p. 304-332.

Research output: Contribution to journalArticle

Gregor, Jens ; Whitaker, R. T. / Indoor scene reconstruction from sets of noisy range images. In: Graphical Models. 2001 ; Vol. 63, No. 5. pp. 304-332.
@article{0440577fa4784ba08378cba564ffda35,
title = "Indoor scene reconstruction from sets of noisy range images",
abstract = "This paper describes a system for building 3D models of indoor scenes from sets of noisy laser range images. It addresses several important aspects of this problem, namely, preprocessing, which includes image segmentation and planar model fitting; view registration, which is the method of determining the rigid transformation that describes the relative pose of the camera platform between views; and reconstruction, which is the subsequent integration or fusion of separate range images into a single 3D model. Our proposed strategy is to use a statistical sensor model. We thus account for noise properties of the data at each stage in the reconstruction process, which produces reliable results even in presence of significant measurement noise. We given a empirical analysis of a plane-based registration method and present results using real range data that demonstrate the performance of the entire reconstruction system.",
author = "Jens Gregor and Whitaker, {R. T.}",
year = "2001",
month = "1",
day = "1",
doi = "10.1006/gmod.2001.0562",
language = "English (US)",
volume = "63",
pages = "304--332",
journal = "Graphical Models",
issn = "1524-0703",
publisher = "Elsevier Inc.",
number = "5",

}

TY - JOUR

T1 - Indoor scene reconstruction from sets of noisy range images

AU - Gregor, Jens

AU - Whitaker, R. T.

PY - 2001/1/1

Y1 - 2001/1/1

N2 - This paper describes a system for building 3D models of indoor scenes from sets of noisy laser range images. It addresses several important aspects of this problem, namely, preprocessing, which includes image segmentation and planar model fitting; view registration, which is the method of determining the rigid transformation that describes the relative pose of the camera platform between views; and reconstruction, which is the subsequent integration or fusion of separate range images into a single 3D model. Our proposed strategy is to use a statistical sensor model. We thus account for noise properties of the data at each stage in the reconstruction process, which produces reliable results even in presence of significant measurement noise. We given a empirical analysis of a plane-based registration method and present results using real range data that demonstrate the performance of the entire reconstruction system.

AB - This paper describes a system for building 3D models of indoor scenes from sets of noisy laser range images. It addresses several important aspects of this problem, namely, preprocessing, which includes image segmentation and planar model fitting; view registration, which is the method of determining the rigid transformation that describes the relative pose of the camera platform between views; and reconstruction, which is the subsequent integration or fusion of separate range images into a single 3D model. Our proposed strategy is to use a statistical sensor model. We thus account for noise properties of the data at each stage in the reconstruction process, which produces reliable results even in presence of significant measurement noise. We given a empirical analysis of a plane-based registration method and present results using real range data that demonstrate the performance of the entire reconstruction system.

UR - http://www.scopus.com/inward/record.url?scp=0035467054&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035467054&partnerID=8YFLogxK

U2 - 10.1006/gmod.2001.0562

DO - 10.1006/gmod.2001.0562

M3 - Article

VL - 63

SP - 304

EP - 332

JO - Graphical Models

JF - Graphical Models

SN - 1524-0703

IS - 5

ER -