Learning to estimate human pose with data driven belief propagation

Gang Hua, Ming Hsuan Yang, Ying Wu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

85 Citations (Scopus)

Abstract

We propose a statistical formulation for 2-D human pose estimation from, single images. The human body configuration is modeled by a Markov network and the estimation problem is to infer pose parameters from image cues such as appearance, shape, edge, and color. From a set of hand labeled images, we accumulate prior knowledge of 2-D body shapes by learning their low-dimensional representations for inference of pose parameters. A data driven belief propagation Monte Carlo algorithm, utilizing importance sampling functions built from bottom-up visual cues, is proposed for efficient probabilistic inference. Contrasted to the few sequential statistical formulations in the literature, our algorithm integrates both top-down as well as bottom-up reasoning mechanisms, and can carry out the inference tasks in parallel. Experimental results demonstrate the potency and effectiveness of the proposed algorithm in estimating 2-D human pose from single images.

Original languageEnglish
Title of host publicationProceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005
PublisherIEEE Computer Society
Pages747-754
Number of pages8
ISBN (Print)0769523722, 9780769523729
DOIs
Publication statusPublished - 2005 Jan 1
Event2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005 - San Diego, CA, United States
Duration: 2005 Jun 202005 Jun 25

Publication series

NameProceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005
VolumeII

Conference

Conference2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005
CountryUnited States
CitySan Diego, CA
Period05/6/2005/6/25

Fingerprint

Importance sampling
Color

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Hua, G., Yang, M. H., & Wu, Y. (2005). Learning to estimate human pose with data driven belief propagation. In Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005 (pp. 747-754). [1467517] (Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005; Vol. II). IEEE Computer Society. https://doi.org/10.1109/CVPR.2005.208
Hua, Gang ; Yang, Ming Hsuan ; Wu, Ying. / Learning to estimate human pose with data driven belief propagation. Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005. IEEE Computer Society, 2005. pp. 747-754 (Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005).
@inproceedings{c5ce47b081444461adcd6b5a7e5430ee,
title = "Learning to estimate human pose with data driven belief propagation",
abstract = "We propose a statistical formulation for 2-D human pose estimation from, single images. The human body configuration is modeled by a Markov network and the estimation problem is to infer pose parameters from image cues such as appearance, shape, edge, and color. From a set of hand labeled images, we accumulate prior knowledge of 2-D body shapes by learning their low-dimensional representations for inference of pose parameters. A data driven belief propagation Monte Carlo algorithm, utilizing importance sampling functions built from bottom-up visual cues, is proposed for efficient probabilistic inference. Contrasted to the few sequential statistical formulations in the literature, our algorithm integrates both top-down as well as bottom-up reasoning mechanisms, and can carry out the inference tasks in parallel. Experimental results demonstrate the potency and effectiveness of the proposed algorithm in estimating 2-D human pose from single images.",
author = "Gang Hua and Yang, {Ming Hsuan} and Ying Wu",
year = "2005",
month = "1",
day = "1",
doi = "10.1109/CVPR.2005.208",
language = "English",
isbn = "0769523722",
series = "Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005",
publisher = "IEEE Computer Society",
pages = "747--754",
booktitle = "Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005",
address = "United States",

}

Hua, G, Yang, MH & Wu, Y 2005, Learning to estimate human pose with data driven belief propagation. in Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005., 1467517, Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, vol. II, IEEE Computer Society, pp. 747-754, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, San Diego, CA, United States, 05/6/20. https://doi.org/10.1109/CVPR.2005.208

Learning to estimate human pose with data driven belief propagation. / Hua, Gang; Yang, Ming Hsuan; Wu, Ying.

Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005. IEEE Computer Society, 2005. p. 747-754 1467517 (Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005; Vol. II).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Learning to estimate human pose with data driven belief propagation

AU - Hua, Gang

AU - Yang, Ming Hsuan

AU - Wu, Ying

PY - 2005/1/1

Y1 - 2005/1/1

N2 - We propose a statistical formulation for 2-D human pose estimation from, single images. The human body configuration is modeled by a Markov network and the estimation problem is to infer pose parameters from image cues such as appearance, shape, edge, and color. From a set of hand labeled images, we accumulate prior knowledge of 2-D body shapes by learning their low-dimensional representations for inference of pose parameters. A data driven belief propagation Monte Carlo algorithm, utilizing importance sampling functions built from bottom-up visual cues, is proposed for efficient probabilistic inference. Contrasted to the few sequential statistical formulations in the literature, our algorithm integrates both top-down as well as bottom-up reasoning mechanisms, and can carry out the inference tasks in parallel. Experimental results demonstrate the potency and effectiveness of the proposed algorithm in estimating 2-D human pose from single images.

AB - We propose a statistical formulation for 2-D human pose estimation from, single images. The human body configuration is modeled by a Markov network and the estimation problem is to infer pose parameters from image cues such as appearance, shape, edge, and color. From a set of hand labeled images, we accumulate prior knowledge of 2-D body shapes by learning their low-dimensional representations for inference of pose parameters. A data driven belief propagation Monte Carlo algorithm, utilizing importance sampling functions built from bottom-up visual cues, is proposed for efficient probabilistic inference. Contrasted to the few sequential statistical formulations in the literature, our algorithm integrates both top-down as well as bottom-up reasoning mechanisms, and can carry out the inference tasks in parallel. Experimental results demonstrate the potency and effectiveness of the proposed algorithm in estimating 2-D human pose from single images.

UR - http://www.scopus.com/inward/record.url?scp=24644471590&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=24644471590&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2005.208

DO - 10.1109/CVPR.2005.208

M3 - Conference contribution

AN - SCOPUS:24644471590

SN - 0769523722

SN - 9780769523729

T3 - Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005

SP - 747

EP - 754

BT - Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005

PB - IEEE Computer Society

ER -

Hua G, Yang MH, Wu Y. Learning to estimate human pose with data driven belief propagation. In Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005. IEEE Computer Society. 2005. p. 747-754. 1467517. (Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005). https://doi.org/10.1109/CVPR.2005.208