Multi-modal abnormality detection in video with unknown data segmentation

This paper examines a new problem in large scale stream data: abnormality detection which is localized to a data segmentation process. Unlike traditional abnormality detection methods which typically build one unified model across data stream, we propose that building multiple detection models focus...

Full description

Bibliographic Details
Main Authors: Nguyen, T., Rana, S., Phung, D., Pham, DucSon, Venkatesh, S.
Other Authors: N/A
Format: Conference Paper
Published: IEEE 2012
Subjects:
Online Access:http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6460383
http://hdl.handle.net/20.500.11937/42528
_version_ 1848756444107636736
author Nguyen, T.
Rana, S.
Phung, D.
Pham, DucSon
Venkatesh, S.
author2 N/A
author_facet N/A
Nguyen, T.
Rana, S.
Phung, D.
Pham, DucSon
Venkatesh, S.
author_sort Nguyen, T.
building Curtin Institutional Repository
collection Online Access
description This paper examines a new problem in large scale stream data: abnormality detection which is localized to a data segmentation process. Unlike traditional abnormality detection methods which typically build one unified model across data stream, we propose that building multiple detection models focused on different coherent sections of the video stream would result in better detection performance. One key challenge is to segment the data into coherent sections as the number of segments is not known in advance and can vary greatly across cameras; and a principled way approach is required. To this end, we first employ the recently pro-posed infinite HMM and collapsed Gibbs inference to automatically infer data segmentation followed by constructing abnormality detection models which are localized to each segmentation. We demonstrate the superior performance of the proposed framework in a real-world surveillance camera data over 14 days.
first_indexed 2025-11-14T09:12:17Z
format Conference Paper
id curtin-20.500.11937-42528
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T09:12:17Z
publishDate 2012
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-425282017-01-30T15:00:08Z Multi-modal abnormality detection in video with unknown data segmentation Nguyen, T. Rana, S. Phung, D. Pham, DucSon Venkatesh, S. N/A vectors detectors surveillance hidden Markov models computational modeling data models cameras This paper examines a new problem in large scale stream data: abnormality detection which is localized to a data segmentation process. Unlike traditional abnormality detection methods which typically build one unified model across data stream, we propose that building multiple detection models focused on different coherent sections of the video stream would result in better detection performance. One key challenge is to segment the data into coherent sections as the number of segments is not known in advance and can vary greatly across cameras; and a principled way approach is required. To this end, we first employ the recently pro-posed infinite HMM and collapsed Gibbs inference to automatically infer data segmentation followed by constructing abnormality detection models which are localized to each segmentation. We demonstrate the superior performance of the proposed framework in a real-world surveillance camera data over 14 days. 2012 Conference Paper http://hdl.handle.net/20.500.11937/42528 http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6460383 IEEE restricted
spellingShingle vectors
detectors
surveillance
hidden Markov models
computational modeling
data models
cameras
Nguyen, T.
Rana, S.
Phung, D.
Pham, DucSon
Venkatesh, S.
Multi-modal abnormality detection in video with unknown data segmentation
title Multi-modal abnormality detection in video with unknown data segmentation
title_full Multi-modal abnormality detection in video with unknown data segmentation
title_fullStr Multi-modal abnormality detection in video with unknown data segmentation
title_full_unstemmed Multi-modal abnormality detection in video with unknown data segmentation
title_short Multi-modal abnormality detection in video with unknown data segmentation
title_sort multi-modal abnormality detection in video with unknown data segmentation
topic vectors
detectors
surveillance
hidden Markov models
computational modeling
data models
cameras
url http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6460383
http://hdl.handle.net/20.500.11937/42528