DMCA
Probabilistic Visual Learning for Object Representation (1996)
Cached
Download Links
- [ftp.cs.columbia.edu]
- [whitechapel.media.mit.edu]
- DBLP
Other Repositories/Bibliography
| Citations: | 699 - 15 self |
BibTeX
@MISC{Moghaddam96probabilisticvisual,
author = {Baback Moghaddam and Alex Pentland},
title = {Probabilistic Visual Learning for Object Representation},
year = {1996}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We present an unsupervised technique for visual learning which is based on density estimation in high-dimensional spaces using an eigenspace decomposition. Two types of density estimates are derived for modeling the training data: a multivariate Gaussian (for unimodal distributions) and a Mixture-of-Gaussians model (for multimodal distributions). These probability densities are then used to formulate a maximum-likelihood estimation framework for visual search and target detection for automatic object recognition and coding. Our learning technique is applied to the probabilistic visual modeling, detection, recognition, and coding of human faces and non-rigid objects such as hands.
Keyphrases
probabilistic visual learning object representation density estimate training data high-dimensional space visual search density estimation multimodal distribution mixture-of-gaussians model probabilistic visual modeling multivariate gaussian maximum-likelihood estimation framework non-rigid object automatic object recognition unimodal distribution learning technique probability density eigenspace decomposition visual learning target detection unsupervised technique
