Recognition and Reproduction of Gestures Using a Probabilistic Framework Combining PCA, ICA and HMM
Abstract
This paper explores the issue of recognizing, generalizing and reproducing arbitrary gestures. We aim at extracting a representation that encapsulates only the key aspects of the gesture and discards the variability intrinsic to each person's motion. We compare a decomposition into principal components (PCA) and independent components (ICA) as a first step of preprocessing in order to decorrelate and denoise the data, as well as to reduce the dimensionality of the dataset to make this one tractable. In a second stage of processing, we explore the use of a probabilistic encoding through continuous Hidden Markov Models (HMMs), as a way to encapsulate the sequential nature and intrinsic variability of the motions in stochastic finite state automata. Finally, the method is validated in a humanoid robot to reproduce a variety of gestures performed by a human demonstrator.
Cite
Text
Calinon and Billard. "Recognition and Reproduction of Gestures Using a Probabilistic Framework Combining PCA, ICA and HMM." International Conference on Machine Learning, 2005. doi:10.1145/1102351.1102365Markdown
[Calinon and Billard. "Recognition and Reproduction of Gestures Using a Probabilistic Framework Combining PCA, ICA and HMM." International Conference on Machine Learning, 2005.](https://mlanthology.org/icml/2005/calinon2005icml-recognition/) doi:10.1145/1102351.1102365BibTeX
@inproceedings{calinon2005icml-recognition,
title = {{Recognition and Reproduction of Gestures Using a Probabilistic Framework Combining PCA, ICA and HMM}},
author = {Calinon, Sylvain and Billard, Aude},
booktitle = {International Conference on Machine Learning},
year = {2005},
pages = {105-112},
doi = {10.1145/1102351.1102365},
url = {https://mlanthology.org/icml/2005/calinon2005icml-recognition/}
}