%Aigaion2 BibTeX export from Idiap Publications %Thursday 21 November 2024 01:14:31 PM @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}, }