Generalised Pose Estimation Using Depth (bibtex)

by Simon Hadfield and Richard Bowden

Abstract:

Estimating the pose of an object, be it articulated, deformable or rigid, is an important task, with applications ranging from Human-Computer Interaction to environmental understanding. The idea of a general pose estimation framework, capable of being rapidly retrained to suit a variety of tasks, is appealing. In this paper a solution isproposed requiring only a set of labelled training images in order to be applied to many pose estimation tasks. This is achieved bytreating pose estimation as a classification problem, with particle filtering used to provide non-discretised estimates. Depth information extracted from a calibrated stereo sequence, is used for background suppression and object scale estimation. The appearance and shape channels are then transformed to Local Binary Pattern histograms, and pose classification is performed via a randomised decision forest. To demonstrate flexibility, the approach is applied to two different situations, articulated hand pose and rigid head orientation, achieving 97% and 84% accurate estimation rates, respectively.

View PDF

Reference:

Generalised Pose Estimation Using Depth (Simon Hadfield and Richard Bowden), In Proceedings, International Workshop on Sign, Gesture and Activity at ECCV 2010 (Kiriakos N. Kutulakos, ed.), Springer, volume 6553, 2010. (Poster)

Bibtex Entry:

@InProceedings{Hadfield10,
  Title                    = {Generalised Pose Estimation Using Depth},
  Author                   = {Simon Hadfield and Richard Bowden},
  Booktitle                = {Proceedings, International Workshop on Sign, Gesture and Activity at ECCV 2010},
  Year                     = {2010},

  Address                  = {Heraklion, Crete},
  Editor                   = {Kiriakos N. Kutulakos},
  Month                    = {5 -- 11 } # sep,
  Pages                    = {312 -- 325},
  Publisher                = {Springer},
  Series                   = {Lecture Notes in Computer Science},
  Volume                   = {6553},

  Abstract                 = {Estimating the pose of an object, be it articulated, deformable or rigid, is an important task, with applications ranging from Human-Computer Interaction to environmental understanding. The idea of a general pose estimation framework, capable of being rapidly retrained to suit a variety of tasks, is appealing. In this paper a solution isproposed requiring only a set of labelled training images in order to be applied to many pose estimation tasks. This is achieved bytreating pose estimation as a classification problem, with particle filtering used to provide non-discretised estimates. Depth information extracted from a calibrated stereo sequence, is used for background suppression and object scale estimation. The appearance and shape channels are then transformed to Local Binary Pattern histograms, and pose classification is performed via a randomised decision forest. To demonstrate flexibility, the approach is applied to two different situations, articulated hand pose and rigid head orientation, achieving 97% and 84% accurate estimation rates, respectively.},
  Comment                  = {<a href="http://personalpages.surrey.ac.uk/s.hadfield/posters/Generalised%20Pose%20Estimation%20Using%20Depth.tif">Poster</a>},
  Crossref                 = {ECCV10},
  Doi                      = {10.1007/978-3-642-35749-7_24},
  File                     = {Hadfield10.pdf:Hadfield10.pdf:PDF},
  Gsid                     = {6786518533375756538,12665075001326605511},
  Keywords                 = {pose, depth, stereo, head, hand, classification, particle filter, gesture, lbp, rdf, background suppression, object extraction, segmentation},
  Timestamp                = {2010.09.14},
  Url                      = {http://personalpages.surrey.ac.uk/s.hadfield/papers/Generalised%20Pose%20Estimation%20Using%20Depth.pdf}
}