Shape from photographs: a multi-view stereo pipeline

Carlos Hernández; George Vogiatzis

doi:10.1007/978-3-642-12848-6_11

Shape from photographs: a multi-view stereo pipeline

Carlos Hernández^*, George Vogiatzis

^*Corresponding author for this work

Computer Science Research Group

Toshiba Research Cambridge

Research output: Chapter in Book/Published conference output › Chapter

16 Citations (Scopus)

Abstract

Acquiring 3D shape from images is a classic problem in Computer Vision occupying researchers for at least 20 years. Only recently however have these ideas matured enough to provide highly accurate results. We present a complete algorithm to reconstruct 3D objects from images using the stereo correspondence cue. The technique can be described as a pipeline of four basic building blocks: camera calibration, image segmentation, photo-consistency estimation from images, and surface extraction from photo-consistency. In this Chapter we will put more emphasis on the latter two: namely how to extract geometric information from a set of photographs without explicit camera visibility, and how to combine different geometry estimates in an optimal way.

Original language	English
Title of host publication	Computer vision
Subtitle of host publication	detection, recognition and reconstruction
Editors	Roberto Cipolla, Sebastiano Battiato, Giovanni Maria Farinella
Place of Publication	Berlin (US)
Publisher	Springer
Pages	281-311
Number of pages	31
ISBN (Electronic)	978-3-642-12848-6
ISBN (Print)	978-3-642-12847-9
DOIs	https://doi.org/10.1007/978-3-642-12848-6_11
Publication status	Published - 2010

Publication series

Name	Studies in Computational Intelligence
Publisher	Springer
Volume	285
ISSN (Print)	1860-949X

Access to Document

10.1007/978-3-642-12848-6_11

Cite this

@inbook{80e62942b0ed44f5a98fefc52c467cf1,

title = "Shape from photographs: a multi-view stereo pipeline",

abstract = "Acquiring 3D shape from images is a classic problem in Computer Vision occupying researchers for at least 20 years. Only recently however have these ideas matured enough to provide highly accurate results. We present a complete algorithm to reconstruct 3D objects from images using the stereo correspondence cue. The technique can be described as a pipeline of four basic building blocks: camera calibration, image segmentation, photo-consistency estimation from images, and surface extraction from photo-consistency. In this Chapter we will put more emphasis on the latter two: namely how to extract geometric information from a set of photographs without explicit camera visibility, and how to combine different geometry estimates in an optimal way.",

author = "Carlos Hern{\'a}ndez and George Vogiatzis",

year = "2010",

doi = "10.1007/978-3-642-12848-6_11",

language = "English",

isbn = "978-3-642-12847-9",

series = "Studies in Computational Intelligence",

publisher = "Springer",

pages = "281--311",

editor = "Roberto Cipolla and Sebastiano Battiato and Farinella, {Giovanni Maria}",

booktitle = "Computer vision",

address = "Germany",

}

Shape from photographs: a multi-view stereo pipeline. / Hernández, Carlos; Vogiatzis, George.
Computer vision: detection, recognition and reconstruction. ed. / Roberto Cipolla; Sebastiano Battiato; Giovanni Maria Farinella. Berlin (US): Springer, 2010. p. 281-311 (Studies in Computational Intelligence; Vol. 285).

Research output: Chapter in Book/Published conference output › Chapter

TY - CHAP

T1 - Shape from photographs

T2 - a multi-view stereo pipeline

AU - Hernández, Carlos

AU - Vogiatzis, George

PY - 2010

Y1 - 2010

N2 - Acquiring 3D shape from images is a classic problem in Computer Vision occupying researchers for at least 20 years. Only recently however have these ideas matured enough to provide highly accurate results. We present a complete algorithm to reconstruct 3D objects from images using the stereo correspondence cue. The technique can be described as a pipeline of four basic building blocks: camera calibration, image segmentation, photo-consistency estimation from images, and surface extraction from photo-consistency. In this Chapter we will put more emphasis on the latter two: namely how to extract geometric information from a set of photographs without explicit camera visibility, and how to combine different geometry estimates in an optimal way.

AB - Acquiring 3D shape from images is a classic problem in Computer Vision occupying researchers for at least 20 years. Only recently however have these ideas matured enough to provide highly accurate results. We present a complete algorithm to reconstruct 3D objects from images using the stereo correspondence cue. The technique can be described as a pipeline of four basic building blocks: camera calibration, image segmentation, photo-consistency estimation from images, and surface extraction from photo-consistency. In this Chapter we will put more emphasis on the latter two: namely how to extract geometric information from a set of photographs without explicit camera visibility, and how to combine different geometry estimates in an optimal way.

UR - http://www.scopus.com/inward/record.url?scp=77950639127&partnerID=8YFLogxK

UR - http://link.springer.com/chapter/10.1007%2F978-3-642-12848-6_11

U2 - 10.1007/978-3-642-12848-6_11

DO - 10.1007/978-3-642-12848-6_11

M3 - Chapter

AN - SCOPUS:77950639127

SN - 978-3-642-12847-9

T3 - Studies in Computational Intelligence

SP - 281

EP - 311

BT - Computer vision

A2 - Cipolla, Roberto

A2 - Battiato, Sebastiano

A2 - Farinella, Giovanni Maria

PB - Springer

CY - Berlin (US)

ER -

Shape from photographs: a multi-view stereo pipeline

Abstract

Publication series

Access to Document

Other files and links

Fingerprint

Practical 3D reconstruction based on photometric stereo

Cite this