Priors for People Tracking from Small Training Sets
Abstract
We advocate the use of scaled Gaussian process latent variable models (SGPLVM) to learn prior models of 3D human pose for 3D people tracking. The SGPLVM simultaneously optimizes a low-dimensional embedding of the high-dimensional pose data and a density function that both gives higher probability to points close to training data and provides a nonlinear probabilistic mapping from the low-dimensional latent space to the full-dimensional pose space. The SGPLVM is a natural choice when only small amounts of training data are available. We demonstrate our approach with two distinct motions, golfing and walking. We show that the SGPLVM sufficiently constrains the problem such that tracking can be accomplished with straightforward deterministic optimization.
Cite
Text
Urtasun et al. "Priors for People Tracking from Small Training Sets." IEEE/CVF International Conference on Computer Vision, 2005. doi:10.1109/ICCV.2005.193Markdown
[Urtasun et al. "Priors for People Tracking from Small Training Sets." IEEE/CVF International Conference on Computer Vision, 2005.](https://mlanthology.org/iccv/2005/urtasun2005iccv-priors/) doi:10.1109/ICCV.2005.193BibTeX
@inproceedings{urtasun2005iccv-priors,
title = {{Priors for People Tracking from Small Training Sets}},
author = {Urtasun, Raquel and Fleet, David J. and Hertzmann, Aaron and Fua, Pascal},
booktitle = {IEEE/CVF International Conference on Computer Vision},
year = {2005},
pages = {403-410},
doi = {10.1109/ICCV.2005.193},
url = {https://mlanthology.org/iccv/2005/urtasun2005iccv-priors/}
}