Hidden Markov models with kernel density estimation of emission probabilities and their use in activity recognition

Piccardi, M; Pérez, O

Hidden Markov models with kernel density estimation of emission probabilities and their use in activity recognition

Piccardi, M

Pérez, O

Permalink

Publication Type:: Conference Proceeding
Citation:: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2007
Issue Date:: 2007-10-11

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download full textAdobe PDF (315.9 kB)

View on publisher's site

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Piccardi, M https://orcid.org/0000-0001-9250-6604	en_US
dc.contributor.author	Pérez, O	en_US
dc.date.issued	2007-10-11	en_US
dc.identifier.citation	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2007	en_US
dc.identifier.isbn	1424411807	en_US
dc.identifier.isbn	9781424411801	en_US
dc.identifier.issn	1063-6919	en_US
dc.identifier.uri	http://hdl.handle.net/10453/2542
dc.description.abstract	In this paper, we present a modified hidden Markov model with emission probabilities modelled by kernel density estimation and its use for activity recognition in videos. In the proposed approach, kernel density estimation of the emission probabilities is operated simultaneously with that of all the other model parameters by an adapted Baum-Welch algorithm. This allows us to retain maximum-likelihood estimation while overcoming the known limitations of mixture of Gaussions in modelling certain probability distributions. Experiments on activity recognition have been performed on groundtruthed data from the CAVIAR video surveillance database and reported in the paper. The error on the training and validation sets with kernel density estimation remains around 14-16% while for the conventional Gaussian mixture approach varies between 15 and 24%, strongly depending on the initial values chosen for the parameters. Overall, kernel density estimation proves capable of providing more flexible modelling of the emission probabilities and, unlike Gaussian mixtures, does not suffer from being highly parametric and of difficult initialisation. © 2007 IEEE.	en_US
dc.relation.ispartof	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition	en_US
dc.relation.isbasedon	10.1109/CVPR.2007.383504	en_US
dc.title	Hidden Markov models with kernel density estimation of emission probabilities and their use in activity recognition	en_US
dc.type	Conference Proceeding
utslib.for	080104 Computer Vision	en_US
utslib.for	080109 Pattern Recognition and Data Mining	en_US
dc.location.activity	Minneapolis	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Electrical and Data Engineering
pubs.organisational-group	/University of Technology Sydney/Strength - GBDTC - Global Big Data Technologies
utslib.copyright.status	open_access
pubs.publication-status	Published	en_US

Abstract:

In this paper, we present a modified hidden Markov model with emission probabilities modelled by kernel density estimation and its use for activity recognition in videos. In the proposed approach, kernel density estimation of the emission probabilities is operated simultaneously with that of all the other model parameters by an adapted Baum-Welch algorithm. This allows us to retain maximum-likelihood estimation while overcoming the known limitations of mixture of Gaussions in modelling certain probability distributions. Experiments on activity recognition have been performed on groundtruthed data from the CAVIAR video surveillance database and reported in the paper. The error on the training and validation sets with kernel density estimation remains around 14-16% while for the conventional Gaussian mixture approach varies between 15 and 24%, strongly depending on the initial values chosen for the parameters. Overall, kernel density estimation proves capable of providing more flexible modelling of the emission probabilities and, unlike Gaussian mixtures, does not suffer from being highly parametric and of difficult initialisation. © 2007 IEEE.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/2542