International Journal of Computer Vision

, Volume 113, Issue 3, pp 163–175

Metric Regression Forests for Correspondence Estimation

  • Gerard Pons-Moll
  • Jonathan Taylor
  • Jamie Shotton
  • Aaron Hertzmann
  • Andrew Fitzgibbon
Article

DOI: 10.1007/s11263-015-0818-9

Cite this article as:
Pons-Moll, G., Taylor, J., Shotton, J. et al. Int J Comput Vis (2015) 113: 163. doi:10.1007/s11263-015-0818-9
  • 540 Downloads

Abstract

We present a new method for inferring dense data to model correspondences, focusing on the application of human pose estimation from depth images. Recent work proposed the use of regression forests to quickly predict correspondences between depth pixels and points on a 3D human mesh model. That work, however, used a proxy forest training objective based on the classification of depth pixels to body parts. In contrast, we introduce Metric Space Information Gain (MSIG), a new decision forest training objective designed to directly minimize the entropy of distributions in a metric space. When applied to a model surface, viewed as a metric space defined by geodesic distances, MSIG aims to minimize image-to-model correspondence uncertainty. A naïve implementation of MSIG would scale quadratically with the number of training examples. As this is intractable for large datasets, we propose a method to compute MSIG in linear time. Our method is a principled generalization of the proxy classification objective, and does not require an extrinsic isometric embedding of the model surface in Euclidean space. Our experiments demonstrate that this leads to correspondences that are considerably more accurate than state of the art, using far fewer training images.

Keywords

Human pose estimation Model based pose estimation Correspondence estimation Depth images Metric regression forests 

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  • Gerard Pons-Moll
    • 1
  • Jonathan Taylor
    • 2
  • Jamie Shotton
    • 2
  • Aaron Hertzmann
    • 3
  • Andrew Fitzgibbon
    • 2
  1. 1.Max Planck for Intelligent SystemsTübingenGermany
  2. 2.Microsoft ResearchCambridgeUK
  3. 3.Adobe ResearchSan FranciscoUSA

Personalised recommendations