%Aigaion2 BibTeX export from Idiap Publications
%Sunday 22 December 2024 04:05:24 AM

@INPROCEEDINGS{zhang-rr-04-28b,
         author = {Zhang, Dong and Gatica-Perez, Daniel and Bengio, Samy and McCowan, Iain A.},
       projects = {Idiap},
          title = {Semi-supervised Adapted HMMs for Unusual Event Detection},
      booktitle = {Pro. IEEE CVPR},
           year = {2005},
           note = {IDIAP-RR 04-80},
       crossref = {zhang-rr-04-28},
       abstract = {We address the problem of temporal unusual event detection. Unusual events are characterized by a number of features (rarity, unexpectedness, and relevance) that limit the application of traditional supervised model-based approaches. We propose a semi-supervised adapted Hidden Markov Model (HMM) framework, in which usual event models are first learned from a large amount of (commonly available) training data, while unusual event models are learned by Bayesian adaptation in an unsupervised manner. The proposed framework has an iterative structure, which adapts a new unusual event model at each iteration. We show that such a framework can address problems due to the scarcity of training data and the difficulty in pre-defining unusual events. Experiments on audio, visual, and audio-visual data streams illustrate its effectiveness, compared with both supervised and unsupervised baseline methods.},
            pdf = {https://publications.idiap.ch/attachments/reports/2005/zhang-cvpr-05.pdf},
     postscript = {ftp://ftp.idiap.ch/pub/reports/2004/rr-04-80.ps.gz},
ipdmembership={vision},
}



crossreferenced publications: 
@TECHREPORT{zhang-rr-04-28,
         author = {Zhang, Dong and Gatica-Perez, Daniel and Bengio, Samy},
       projects = {Idiap},
          title = {{Semi-supervised Adapted HMMs for Unusual Event Detection}},
           type = {Idiap-RR},
         number = {Idiap-RR-80-2004},
           year = {2004},
    institution = {IDIAP},
        address = {Martigny, Switzerland},
           note = {Published in ``Prof. IEEE CVPR'', June, 2005},
       abstract = {We address the problem of temporal unusual event detection. Unusual events are characterized by a number of features (rarity, unexpectedness, and relevance) that limit the application of traditional supervised model-based approaches. We propose a semi-supervised adapted Hidden Markov Model (HMM) framework, in which usual event models are first learned from a large amount of (commonly available) training data, while unusual event models are learned by Bayesian adaptation in an unsupervised manner. The proposed framework has an iterative structure, which adapts a new unusual event model at each iteration. We show that such a framework can address problems due to the scarcity of training data and the difficulty in pre-defining unusual events. Experiments on audio, visual, and audio-visual data streams illustrate its effectiveness, compared with both supervised and unsupervised baseline methods.},
            pdf = {https://publications.idiap.ch/attachments/reports/2004/rr-04-80.pdf},
     postscript = {ftp://ftp.idiap.ch/pub/reports/2004/rr-04-80.ps.gz},
ipdinar={2004},
ipdmembership={vision},
language={English},
}