CaltechAUTHORS
  A Caltech Library Service

Merging pose estimates across space and time

Burgos-Artizzu, Xavier P. and Hall, David C. and Perona, Pietro and Dollár, Piotr (2013) Merging pose estimates across space and time. In: Proceedings British Machine Vision Conference 2013. Durham, UK , BMVA Press, Art. No. 58. ISBN 1-901725-49-9. https://resolver.caltech.edu/CaltechAUTHORS:20130930-145804309

[img]
Preview
PDF - Published Version
See Usage Policy.

8Mb
[img] Archive (ZIP) (Supplementary Materials) - Supplemental Material
See Usage Policy.

7Mb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20130930-145804309

Abstract

Numerous ‘non-maximum suppression’ (NMS) post-processing schemes have been proposed for merging multiple independent object detections. We propose a generalization of NMS beyond bounding boxes to merge multiple pose estimates in a single frame. The final estimates are centroids rather than medoids as in standard NMS, thus being more accurate than any of the individual candidates. Using the same mathematical framework, we extend our approach to the multi-frame setting, merging multiple independent pose estimates across space and time and outputting both the number and pose of the objects present in a scene. Our approach sidesteps many of the inherent challenges associated with full tracking (e.g. objects entering/leaving a scene, extended periods of occlusion, etc.). We show its versatility by applying it to two distinct state-of-the-art pose estimation algorithms in three domains: human bodies, faces and mice. Our approach improves both detection accuracy (by helping disambiguate correspondences) as well as pose estimation quality and is computationally efficient.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
http://dx.doi.org/10.5244/C.27.58DOIArticle
ORCID:
AuthorORCID
Perona, Pietro0000-0002-7583-5809
Additional Information:© 2013 The copyright of this document resides with its authors. It may be distributed unchanged freely in print or electronic forms. This work is funded by the Gordon and Betty Moore Foundation and ONR MURI Grant N00014-10-1-0933.
Funders:
Funding AgencyGrant Number
Gordon and Betty Moore FoundationUNSPECIFIED
Office of Naval Research (ONR)N00014-10-1-0933
Record Number:CaltechAUTHORS:20130930-145804309
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20130930-145804309
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:41565
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:30 Sep 2013 22:09
Last Modified:03 Oct 2019 05:50

Repository Staff Only: item control page