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 audiovisual data streams illustrate its effectiveness, compared with both supervised and unsupervised baseline methods.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Semi-supervised adapted HMMs for unusual event detection


    Beteiligte:
    Dong Zhang, (Autor:in) / Gatica-Perez, D. (Autor:in) / Bengio, S. (Autor:in) / McCowan, I. (Autor:in)


    Erscheinungsdatum :

    2005-01-01


    Format / Umfang :

    249461 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    HMMs for Anomaly Detection in Autonomous Robots

    Davide Azzalini / Alberto Castellini / Matteo Luperto et al. | BASE | 2020

    Freier Zugriff

    Semi-Supervised Face Detection

    Sebe, N. / Cohen, I. / Huang, T.S. et al. | IEEE | 2005


    A Comparative Study of Mixture-Gaussian VQ, Ergodic HMMs and Left-to-Right HMMs for Speaker Recognition

    Zhu, X. / Millar, B. / Macleod, I. et al. | British Library Conference Proceedings | 1994


    3D surface analysis using coupled HMMs

    Pernkopf, F. | British Library Online Contents | 2005