Topology free hidden Markov models: application to background modeling. Stenger, B., Ramesh, V., Paragios, N., Coetzee, F., & Buhmann, J. In Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001, volume 1, pages 294–301 vol.1, July, 2001. doi abstract bibtex Hidden Markov models (HMMs) are increasingly being used in computer vision for applications such as: gesture analysis, action recognition from video, and illumination modeling. Their use involves an off-line learning step that is used as a basis for on-line decision making (i.e. a stationarity assumption on the model parameters). But, real-world applications are often non-stationary in nature. This leads to the need for a dynamic mechanism to learn and update the model topology as well as its parameters. This paper presents a new framework for HMM topology and parameter estimation in an online, dynamic fashion. The topology and parameter estimation is posed as a model selection problem with an MDL prior. Online modifications to the topology are made possible by incorporating a state splitting criterion. To demonstrate the potential of the algorithm, the background modeling problem is considered. Theoretical validation and real experiments are presented.
@inproceedings{stenger_topology_2001,
title = {Topology free hidden {Markov} models: application to background modeling},
volume = {1},
shorttitle = {Topology free hidden {Markov} models},
doi = {10.1109/ICCV.2001.937532},
abstract = {Hidden Markov models (HMMs) are increasingly being used in computer vision for applications such as: gesture analysis, action recognition from video, and illumination modeling. Their use involves an off-line learning step that is used as a basis for on-line decision making (i.e. a stationarity assumption on the model parameters). But, real-world applications are often non-stationary in nature. This leads to the need for a dynamic mechanism to learn and update the model topology as well as its parameters. This paper presents a new framework for HMM topology and parameter estimation in an online, dynamic fashion. The topology and parameter estimation is posed as a model selection problem with an MDL prior. Online modifications to the topology are made possible by incorporating a state splitting criterion. To demonstrate the potential of the algorithm, the background modeling problem is considered. Theoretical validation and real experiments are presented.},
booktitle = {Proceedings {Eighth} {IEEE} {International} {Conference} on {Computer} {Vision}. {ICCV} 2001},
author = {Stenger, B. and Ramesh, V. and Paragios, N. and Coetzee, F. and Buhmann, J.M.},
month = jul,
year = {2001},
keywords = {Application software, Computer science, Computer vision, Hidden Markov models, Image analysis, Parameter estimation, Signal processing algorithms, State estimation, Topology, Visualization, hmm, unknown topology},
pages = {294--301 vol.1},
}
Downloads: 0
{"_id":"Ece8ReofR8CFToKJn","bibbaseid":"stenger-ramesh-paragios-coetzee-buhmann-topologyfreehiddenmarkovmodelsapplicationtobackgroundmodeling-2001","author_short":["Stenger, B.","Ramesh, V.","Paragios, N.","Coetzee, F.","Buhmann, J."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Topology free hidden Markov models: application to background modeling","volume":"1","shorttitle":"Topology free hidden Markov models","doi":"10.1109/ICCV.2001.937532","abstract":"Hidden Markov models (HMMs) are increasingly being used in computer vision for applications such as: gesture analysis, action recognition from video, and illumination modeling. Their use involves an off-line learning step that is used as a basis for on-line decision making (i.e. a stationarity assumption on the model parameters). But, real-world applications are often non-stationary in nature. This leads to the need for a dynamic mechanism to learn and update the model topology as well as its parameters. This paper presents a new framework for HMM topology and parameter estimation in an online, dynamic fashion. The topology and parameter estimation is posed as a model selection problem with an MDL prior. Online modifications to the topology are made possible by incorporating a state splitting criterion. To demonstrate the potential of the algorithm, the background modeling problem is considered. Theoretical validation and real experiments are presented.","booktitle":"Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001","author":[{"propositions":[],"lastnames":["Stenger"],"firstnames":["B."],"suffixes":[]},{"propositions":[],"lastnames":["Ramesh"],"firstnames":["V."],"suffixes":[]},{"propositions":[],"lastnames":["Paragios"],"firstnames":["N."],"suffixes":[]},{"propositions":[],"lastnames":["Coetzee"],"firstnames":["F."],"suffixes":[]},{"propositions":[],"lastnames":["Buhmann"],"firstnames":["J.M."],"suffixes":[]}],"month":"July","year":"2001","keywords":"Application software, Computer science, Computer vision, Hidden Markov models, Image analysis, Parameter estimation, Signal processing algorithms, State estimation, Topology, Visualization, hmm, unknown topology","pages":"294–301 vol.1","bibtex":"@inproceedings{stenger_topology_2001,\n\ttitle = {Topology free hidden {Markov} models: application to background modeling},\n\tvolume = {1},\n\tshorttitle = {Topology free hidden {Markov} models},\n\tdoi = {10.1109/ICCV.2001.937532},\n\tabstract = {Hidden Markov models (HMMs) are increasingly being used in computer vision for applications such as: gesture analysis, action recognition from video, and illumination modeling. Their use involves an off-line learning step that is used as a basis for on-line decision making (i.e. a stationarity assumption on the model parameters). But, real-world applications are often non-stationary in nature. This leads to the need for a dynamic mechanism to learn and update the model topology as well as its parameters. This paper presents a new framework for HMM topology and parameter estimation in an online, dynamic fashion. The topology and parameter estimation is posed as a model selection problem with an MDL prior. Online modifications to the topology are made possible by incorporating a state splitting criterion. To demonstrate the potential of the algorithm, the background modeling problem is considered. Theoretical validation and real experiments are presented.},\n\tbooktitle = {Proceedings {Eighth} {IEEE} {International} {Conference} on {Computer} {Vision}. {ICCV} 2001},\n\tauthor = {Stenger, B. and Ramesh, V. and Paragios, N. and Coetzee, F. and Buhmann, J.M.},\n\tmonth = jul,\n\tyear = {2001},\n\tkeywords = {Application software, Computer science, Computer vision, Hidden Markov models, Image analysis, Parameter estimation, Signal processing algorithms, State estimation, Topology, Visualization, hmm, unknown topology},\n\tpages = {294--301 vol.1},\n}\n\n\n\n","author_short":["Stenger, B.","Ramesh, V.","Paragios, N.","Coetzee, F.","Buhmann, J."],"key":"stenger_topology_2001","id":"stenger_topology_2001","bibbaseid":"stenger-ramesh-paragios-coetzee-buhmann-topologyfreehiddenmarkovmodelsapplicationtobackgroundmodeling-2001","role":"author","urls":{},"keyword":["Application software","Computer science","Computer vision","Hidden Markov models","Image analysis","Parameter estimation","Signal processing algorithms","State estimation","Topology","Visualization","hmm","unknown topology"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/mh_lenguyen","dataSources":["iwKepCrWBps7ojhDx"],"keywords":["application software","computer science","computer vision","hidden markov models","image analysis","parameter estimation","signal processing algorithms","state estimation","topology","visualization","hmm","unknown topology"],"search_terms":["topology","free","hidden","markov","models","application","background","modeling","stenger","ramesh","paragios","coetzee","buhmann"],"title":"Topology free hidden Markov models: application to background modeling","year":2001}