Posebits for Monocular Human Pose Estimation

2014

Conference Paper

ps


We advocate the inference of qualitative information about 3D human pose, called posebits, from images. Posebits represent boolean geometric relationships between body parts (e.g., left-leg in front of right-leg or hands close to each other). The advantages of posebits as a mid-level representation are 1) for many tasks of interest, such qualitative pose information may be sufficient (e.g. , semantic image retrieval), 2) it is relatively easy to annotate large image corpora with posebits, as it simply requires answers to yes/no questions; and 3) they help resolve challenging pose ambiguities and therefore facilitate the difficult talk of image-based 3D pose estimation. We introduce posebits, a posebit database, a method for selecting useful posebits for pose estimation and a structural SVM model for posebit inference. Experiments show the use of posebits for semantic image retrieval and for improving 3D pose estimation.

Author(s): Gerard Pons-Moll and David J. Fleet and Bodo Rosenhahn
Book Title: Proceedings IEEE Conf. on Computer Vision and Pattern Recognition (CVPR)
Pages: 2345--2352
Year: 2014
Month: June

Department(s): Perceiving Systems
Research Project(s): 3D Pose from Images
Bibtex Type: Conference Paper (inproceedings)
Paper Type: Conference

Address: Columbus, Ohio, USA
Event Name: IEEE International Conference on Computer Vision and Pattern Recognition
Event Place: Columbus, Ohio, USA
Attachments: pdf

BibTex

@inproceedings{PonsMoll_CVPR2014,
  title = {Posebits for Monocular Human Pose Estimation},
  author = {Pons-Moll, Gerard and Fleet, David J. and Rosenhahn, Bodo},
  booktitle = { Proceedings IEEE Conf. on Computer Vision and Pattern Recognition (CVPR)},
  pages = {2345--2352},
  address = {Columbus, Ohio, USA},
  month = jun,
  year = {2014},
  month_numeric = {6}
}