A generative model for simultaneous estimation of human body shape and pixel-level segmentation

Ingmar Rauschert; Robert T. Collins

doi:10.1007/978-3-642-33715-4_51

A generative model for simultaneous estimation of human body shape and pixel-level segmentation

Ingmar Rauschert, Robert T. Collins

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

15 Scopus citations

Abstract

This paper addresses pixel-level segmentation of a human body from a single image. The problem is formulated as a multi-region segmentation where the human body is constrained to be a collection of geometrically linked regions and the background is split into a small number of distinct zones. We solve this problem in a Bayesian framework for jointly estimating articulated body pose and the pixel-level segmentation of each body part. Using an image likelihood function that simultaneously generates and evaluates the image segmentation corresponding to a given pose, we robustly explore the posterior body shape distribution using a data-driven, coarse-to-fine Metropolis Hastings sampling scheme that includes a strongly data-driven proposal term.

Original language	English (US)
Title of host publication	Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings
Pages	704-717
Number of pages	14
Edition	PART 5
DOIs	https://doi.org/10.1007/978-3-642-33715-4_51
State	Published - 2012
Event	12th European Conference on Computer Vision, ECCV 2012 - Florence, Italy Duration: Oct 7 2012 → Oct 13 2012

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 5
Volume	7576 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	12th European Conference on Computer Vision, ECCV 2012
Country/Territory	Italy
City	Florence
Period	10/7/12 → 10/13/12

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-642-33715-4_51

Cite this

Rauschert, I., & Collins, R. T. (2012). A generative model for simultaneous estimation of human body shape and pixel-level segmentation. In Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings (PART 5 ed., pp. 704-717). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7576 LNCS, No. PART 5). https://doi.org/10.1007/978-3-642-33715-4_51

Rauschert, Ingmar ; Collins, Robert T. / A generative model for simultaneous estimation of human body shape and pixel-level segmentation. Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings. PART 5. ed. 2012. pp. 704-717 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 5).

@inproceedings{5889fdbd802646e9a83e587cb9676f54,

title = "A generative model for simultaneous estimation of human body shape and pixel-level segmentation",

abstract = "This paper addresses pixel-level segmentation of a human body from a single image. The problem is formulated as a multi-region segmentation where the human body is constrained to be a collection of geometrically linked regions and the background is split into a small number of distinct zones. We solve this problem in a Bayesian framework for jointly estimating articulated body pose and the pixel-level segmentation of each body part. Using an image likelihood function that simultaneously generates and evaluates the image segmentation corresponding to a given pose, we robustly explore the posterior body shape distribution using a data-driven, coarse-to-fine Metropolis Hastings sampling scheme that includes a strongly data-driven proposal term.",

author = "Ingmar Rauschert and Collins, {Robert T.}",

year = "2012",

doi = "10.1007/978-3-642-33715-4_51",

language = "English (US)",

isbn = "9783642337147",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

number = "PART 5",

pages = "704--717",

booktitle = "Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings",

edition = "PART 5",

note = "12th European Conference on Computer Vision, ECCV 2012 ; Conference date: 07-10-2012 Through 13-10-2012",

}

Rauschert, I & Collins, RT 2012, A generative model for simultaneous estimation of human body shape and pixel-level segmentation. in Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings. PART 5 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 5, vol. 7576 LNCS, pp. 704-717, 12th European Conference on Computer Vision, ECCV 2012, Florence, Italy, 10/7/12. https://doi.org/10.1007/978-3-642-33715-4_51

A generative model for simultaneous estimation of human body shape and pixel-level segmentation. / Rauschert, Ingmar; Collins, Robert T.
Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings. PART 5. ed. 2012. p. 704-717 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7576 LNCS, No. PART 5).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - A generative model for simultaneous estimation of human body shape and pixel-level segmentation

AU - Rauschert, Ingmar

AU - Collins, Robert T.

PY - 2012

Y1 - 2012

N2 - This paper addresses pixel-level segmentation of a human body from a single image. The problem is formulated as a multi-region segmentation where the human body is constrained to be a collection of geometrically linked regions and the background is split into a small number of distinct zones. We solve this problem in a Bayesian framework for jointly estimating articulated body pose and the pixel-level segmentation of each body part. Using an image likelihood function that simultaneously generates and evaluates the image segmentation corresponding to a given pose, we robustly explore the posterior body shape distribution using a data-driven, coarse-to-fine Metropolis Hastings sampling scheme that includes a strongly data-driven proposal term.

AB - This paper addresses pixel-level segmentation of a human body from a single image. The problem is formulated as a multi-region segmentation where the human body is constrained to be a collection of geometrically linked regions and the background is split into a small number of distinct zones. We solve this problem in a Bayesian framework for jointly estimating articulated body pose and the pixel-level segmentation of each body part. Using an image likelihood function that simultaneously generates and evaluates the image segmentation corresponding to a given pose, we robustly explore the posterior body shape distribution using a data-driven, coarse-to-fine Metropolis Hastings sampling scheme that includes a strongly data-driven proposal term.

UR - http://www.scopus.com/inward/record.url?scp=84867868691&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867868691&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-33715-4_51

DO - 10.1007/978-3-642-33715-4_51

M3 - Conference contribution

AN - SCOPUS:84867868691

SN - 9783642337147

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 704

EP - 717

BT - Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings

T2 - 12th European Conference on Computer Vision, ECCV 2012

Y2 - 7 October 2012 through 13 October 2012

ER -

Rauschert I, Collins RT. A generative model for simultaneous estimation of human body shape and pixel-level segmentation. In Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings. PART 5 ed. 2012. p. 704-717. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 5). doi: 10.1007/978-3-642-33715-4_51

A generative model for simultaneous estimation of human body shape and pixel-level segmentation

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this