Explicit state duration HMM for abnormality detection in sequences of human activity

Much of the current work in human behaviour modelling concentrates on activity recognition, recognising actions and events through pose, movement, and gesture analysis. Our work focuses on learning and detecting abnormality in higher level behavioural patterns. The hidden Markov model (HMM) is one a...

Full description

Bibliographic Details
Main Authors: Luhr, Sebastian, Venkatesh, Svetha, West, Geoffrey, Bui, Hung H.
Other Authors: Chengqi Zhang
Format: Conference Paper
Published: Springer-Verlag 2004
Online Access:http://hdl.handle.net/20.500.11937/8946
_version_ 1848745807850766336
author Luhr, Sebastian
Venkatesh, Svetha
West, Geoffrey
Bui, Hung H.
author2 Chengqi Zhang
author_facet Chengqi Zhang
Luhr, Sebastian
Venkatesh, Svetha
West, Geoffrey
Bui, Hung H.
author_sort Luhr, Sebastian
building Curtin Institutional Repository
collection Online Access
description Much of the current work in human behaviour modelling concentrates on activity recognition, recognising actions and events through pose, movement, and gesture analysis. Our work focuses on learning and detecting abnormality in higher level behavioural patterns. The hidden Markov model (HMM) is one approach for learning such behaviours given a vision tracker recording observations about a persons activity. Duration of human activity is an important consideration if we are to accurately model a persons behavioural patterns. We show how the implicit state duration in the HMM can create a situation in which highly abnormal deviation as either less than or more than the usually observed activity duration can fail to be detected and how the explicit state duration HMM (ESD-HMM) helps alleviate the problem.
first_indexed 2025-11-14T06:23:14Z
format Conference Paper
id curtin-20.500.11937-8946
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T06:23:14Z
publishDate 2004
publisher Springer-Verlag
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-89462022-10-06T07:11:27Z Explicit state duration HMM for abnormality detection in sequences of human activity Luhr, Sebastian Venkatesh, Svetha West, Geoffrey Bui, Hung H. Chengqi Zhang Hans W Guesgen Wai K Yeap Much of the current work in human behaviour modelling concentrates on activity recognition, recognising actions and events through pose, movement, and gesture analysis. Our work focuses on learning and detecting abnormality in higher level behavioural patterns. The hidden Markov model (HMM) is one approach for learning such behaviours given a vision tracker recording observations about a persons activity. Duration of human activity is an important consideration if we are to accurately model a persons behavioural patterns. We show how the implicit state duration in the HMM can create a situation in which highly abnormal deviation as either less than or more than the usually observed activity duration can fail to be detected and how the explicit state duration HMM (ESD-HMM) helps alleviate the problem. 2004 Conference Paper http://hdl.handle.net/20.500.11937/8946 10.1007/978-3-540-28633-2_125 Springer-Verlag restricted
spellingShingle Luhr, Sebastian
Venkatesh, Svetha
West, Geoffrey
Bui, Hung H.
Explicit state duration HMM for abnormality detection in sequences of human activity
title Explicit state duration HMM for abnormality detection in sequences of human activity
title_full Explicit state duration HMM for abnormality detection in sequences of human activity
title_fullStr Explicit state duration HMM for abnormality detection in sequences of human activity
title_full_unstemmed Explicit state duration HMM for abnormality detection in sequences of human activity
title_short Explicit state duration HMM for abnormality detection in sequences of human activity
title_sort explicit state duration hmm for abnormality detection in sequences of human activity
url http://hdl.handle.net/20.500.11937/8946