On detailed 3D reconstruction of large indoor environments

Egor Bondarev

doi:10.1117/12.2085111

On detailed 3D reconstruction of large indoor environments

Egor Bondarev

Architectures for Reliable Image Analysis

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

In this paper we present techniques for highly detailed 3D reconstruction of extra large indoor environments. We discuss the benefits and drawbacks of low-range, far-range and hybrid sensing and reconstruction approaches. The proposed techniques for low-range and hybrid reconstruction, enabling the reconstruction density of 125 points/cm³ on large 100.000 m³ models, are presented in detail. The techniques tackle the core challenges for the above requirements, such as a multi-modal data fusion (fusion of a LIDAR data with a Kinect data), accurate sensor pose estimation, high-density scanning and depth data noise filtering. Other important aspects for extra large 3D indoor reconstruction are the point cloud decimation and real-time rendering. In this paper, we present a method for planar-based point cloud decimation, allowing for reduction of a point cloud size by 80-95%. Besides this, we introduce a method for online rendering of extra large point clouds enabling real-time visualization of huge cloud spaces in conventional web browsers.

Original language	English
Title of host publication	Image Processing: Algorithms and Systems XIII
Editors	K.O. Egiazarian, S.A. Agaian, A.P. Gotchev
Place of Publication	Bellingham
Publisher	SPIE
Number of pages	10
ISBN (Electronic)	9781628414899
DOIs	https://doi.org/10.1117/12.2085111
Publication status	Published - 1 Jan 2015
Event	Image Processing: Algorithms and Systems XIII - San Francisco, United States Duration: 7 Apr 2015 → 7 Apr 2015 http://optics.org/events/2015/477

Publication series

Name	Proceedings of SPIE
Volume	9399
ISSN (Electronic)	0277-786X

Conference

Conference	Image Processing: Algorithms and Systems XIII
Country/Territory	United States
City	San Francisco
Period	7/04/15 → 7/04/15
Other	Part of: 2015 SPIE/IS&T Electronic Imaging Conference
Internet address	http://optics.org/events/2015/477

Keywords

3D reconstruction
Decimation
LIDAR sensing
Mapping and tracking
Point cloud
Real-time
Rendering
RGB-D sensing
Visualization
Volumetric processing

Access to Document

10.1117/12.2085111

Cite this

@inproceedings{4c08cccfeed4421185b46f4c00f1b1d9,

title = "On detailed 3D reconstruction of large indoor environments",

abstract = "In this paper we present techniques for highly detailed 3D reconstruction of extra large indoor environments. We discuss the benefits and drawbacks of low-range, far-range and hybrid sensing and reconstruction approaches. The proposed techniques for low-range and hybrid reconstruction, enabling the reconstruction density of 125 points/cm3 on large 100.000 m3 models, are presented in detail. The techniques tackle the core challenges for the above requirements, such as a multi-modal data fusion (fusion of a LIDAR data with a Kinect data), accurate sensor pose estimation, high-density scanning and depth data noise filtering. Other important aspects for extra large 3D indoor reconstruction are the point cloud decimation and real-time rendering. In this paper, we present a method for planar-based point cloud decimation, allowing for reduction of a point cloud size by 80-95\%. Besides this, we introduce a method for online rendering of extra large point clouds enabling real-time visualization of huge cloud spaces in conventional web browsers.",

keywords = "3D reconstruction, Decimation, LIDAR sensing, Mapping and tracking, Point cloud, Real-time, Rendering, RGB-D sensing, Visualization, Volumetric processing",

author = "Egor Bondarev",

year = "2015",

month = jan,

day = "1",

doi = "10.1117/12.2085111",

language = "English",

series = "Proceedings of SPIE",

publisher = "SPIE",

editor = "K.O. Egiazarian and S.A. Agaian and A.P. Gotchev",

booktitle = "Image Processing: Algorithms and Systems XIII",

address = "United States",

note = "Image Processing: Algorithms and Systems XIII ; Conference date: 07-04-2015 Through 07-04-2015",

url = "http://optics.org/events/2015/477",

}

TY - GEN

T1 - On detailed 3D reconstruction of large indoor environments

AU - Bondarev, Egor

PY - 2015/1/1

Y1 - 2015/1/1

N2 - In this paper we present techniques for highly detailed 3D reconstruction of extra large indoor environments. We discuss the benefits and drawbacks of low-range, far-range and hybrid sensing and reconstruction approaches. The proposed techniques for low-range and hybrid reconstruction, enabling the reconstruction density of 125 points/cm3 on large 100.000 m3 models, are presented in detail. The techniques tackle the core challenges for the above requirements, such as a multi-modal data fusion (fusion of a LIDAR data with a Kinect data), accurate sensor pose estimation, high-density scanning and depth data noise filtering. Other important aspects for extra large 3D indoor reconstruction are the point cloud decimation and real-time rendering. In this paper, we present a method for planar-based point cloud decimation, allowing for reduction of a point cloud size by 80-95%. Besides this, we introduce a method for online rendering of extra large point clouds enabling real-time visualization of huge cloud spaces in conventional web browsers.

AB - In this paper we present techniques for highly detailed 3D reconstruction of extra large indoor environments. We discuss the benefits and drawbacks of low-range, far-range and hybrid sensing and reconstruction approaches. The proposed techniques for low-range and hybrid reconstruction, enabling the reconstruction density of 125 points/cm3 on large 100.000 m3 models, are presented in detail. The techniques tackle the core challenges for the above requirements, such as a multi-modal data fusion (fusion of a LIDAR data with a Kinect data), accurate sensor pose estimation, high-density scanning and depth data noise filtering. Other important aspects for extra large 3D indoor reconstruction are the point cloud decimation and real-time rendering. In this paper, we present a method for planar-based point cloud decimation, allowing for reduction of a point cloud size by 80-95%. Besides this, we introduce a method for online rendering of extra large point clouds enabling real-time visualization of huge cloud spaces in conventional web browsers.

KW - 3D reconstruction

KW - Decimation

KW - LIDAR sensing

KW - Mapping and tracking

KW - Point cloud

KW - Real-time

KW - Rendering

KW - RGB-D sensing

KW - Visualization

KW - Volumetric processing

UR - https://www.scopus.com/pages/publications/84928473716

U2 - 10.1117/12.2085111

DO - 10.1117/12.2085111

M3 - Conference contribution

AN - SCOPUS:84928473716

T3 - Proceedings of SPIE

BT - Image Processing: Algorithms and Systems XIII

A2 - Egiazarian, K.O.

A2 - Agaian, S.A.

A2 - Gotchev, A.P.

PB - SPIE

CY - Bellingham

T2 - Image Processing: Algorithms and Systems XIII

Y2 - 7 April 2015 through 7 April 2015

ER -

On detailed 3D reconstruction of large indoor environments

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this