Learning Temporal Sequence Model from Partially Labeled Data. Shi, Y., Bobick, A., & Essa, I. In Computer Vision and Pattern Recognition, 2006 IEEE Computer Society Conference on, volume 2, pages 1631-1638, 2006. doi abstract bibtex Graphical models are often used to represent and recognize activities. Purely unsupervised methods (such as HMMs) can be trained automatically but yield models whose internal structure - the nodes - are difficult to interpret semantically. Manually constructed networks typically have nodes corresponding to sub-events, but the programming and training of these networks is tedious and requires extensive domain expertise. In this paper, we propose a semi-supervised approach in which a manually structured, Propagation Network (a form of a DBN) is initialized from a small amount of fully annotated data, and then refined by an EM-based learning method in an unsupervised fashion. During node refinement (the M step) a boosting-based algorithm is employed to train the evidence detectors of individual nodes. Experiments on a variety of data types - vision and inertial measurements - in several tasks demonstrate the ability to learn from as little as one fully annotated example accompanied by a small number of positive but non-annotated training examples. The system is applied to both recognition and anomaly detection tasks.
@InProceedings{Shi2006,
Title = {Learning Temporal Sequence Model from Partially Labeled Data},
Author = {Shi, Yifan and Bobick, A. and Essa, I.},
Booktitle = {Computer Vision and Pattern Recognition, 2006 IEEE Computer Society Conference on},
Year = {2006},
Pages = {1631-1638},
Volume = {2},
Abstract = {Graphical models are often used to represent and recognize activities. Purely unsupervised methods (such as HMMs) can be trained automatically but yield models whose internal structure - the nodes - are difficult to interpret semantically. Manually constructed networks typically have nodes corresponding to sub-events, but the programming and training of these networks is tedious and requires extensive domain expertise. In this paper, we propose a semi-supervised approach in which a manually structured, Propagation Network (a form of a DBN) is initialized from a small amount of fully annotated data, and then refined by an EM-based learning method in an unsupervised fashion. During node refinement (the M step) a boosting-based algorithm is employed to train the evidence detectors of individual nodes. Experiments on a variety of data types - vision and inertial measurements - in several tasks demonstrate the ability to learn from as little as one fully annotated example accompanied by a small number of positive but non-annotated training examples. The system is applied to both recognition and anomaly detection tasks.},
Doi = {10.1109/CVPR.2006.174},
ISSN = {1063-6919},
Keywords = {Application software;Boosting;Computer vision;Detectors;Graphical models;Hidden Markov models;Learning systems;Semisupervised learning;State-space methods;Surveillance},
Timestamp = {2014.12.22}
}
Downloads: 0
{"_id":"fq7p5FT7MG6KwiAxS","bibbaseid":"shi-bobick-essa-learningtemporalsequencemodelfrompartiallylabeleddata-2006","downloads":0,"creationDate":"2017-09-14T16:34:37.047Z","title":"Learning Temporal Sequence Model from Partially Labeled Data","author_short":["Shi, Y.","Bobick, A.","Essa, I."],"year":2006,"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/jfslin/jfslin.github.io/master/jf2lin.bib","bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Learning Temporal Sequence Model from Partially Labeled Data","author":[{"propositions":[],"lastnames":["Shi"],"firstnames":["Yifan"],"suffixes":[]},{"propositions":[],"lastnames":["Bobick"],"firstnames":["A."],"suffixes":[]},{"propositions":[],"lastnames":["Essa"],"firstnames":["I."],"suffixes":[]}],"booktitle":"Computer Vision and Pattern Recognition, 2006 IEEE Computer Society Conference on","year":"2006","pages":"1631-1638","volume":"2","abstract":"Graphical models are often used to represent and recognize activities. Purely unsupervised methods (such as HMMs) can be trained automatically but yield models whose internal structure - the nodes - are difficult to interpret semantically. Manually constructed networks typically have nodes corresponding to sub-events, but the programming and training of these networks is tedious and requires extensive domain expertise. In this paper, we propose a semi-supervised approach in which a manually structured, Propagation Network (a form of a DBN) is initialized from a small amount of fully annotated data, and then refined by an EM-based learning method in an unsupervised fashion. During node refinement (the M step) a boosting-based algorithm is employed to train the evidence detectors of individual nodes. Experiments on a variety of data types - vision and inertial measurements - in several tasks demonstrate the ability to learn from as little as one fully annotated example accompanied by a small number of positive but non-annotated training examples. The system is applied to both recognition and anomaly detection tasks.","doi":"10.1109/CVPR.2006.174","issn":"1063-6919","keywords":"Application software;Boosting;Computer vision;Detectors;Graphical models;Hidden Markov models;Learning systems;Semisupervised learning;State-space methods;Surveillance","timestamp":"2014.12.22","bibtex":"@InProceedings{Shi2006,\n Title = {Learning Temporal Sequence Model from Partially Labeled Data},\n Author = {Shi, Yifan and Bobick, A. and Essa, I.},\n Booktitle = {Computer Vision and Pattern Recognition, 2006 IEEE Computer Society Conference on},\n Year = {2006},\n Pages = {1631-1638},\n Volume = {2},\n\n Abstract = {Graphical models are often used to represent and recognize activities. Purely unsupervised methods (such as HMMs) can be trained automatically but yield models whose internal structure - the nodes - are difficult to interpret semantically. Manually constructed networks typically have nodes corresponding to sub-events, but the programming and training of these networks is tedious and requires extensive domain expertise. In this paper, we propose a semi-supervised approach in which a manually structured, Propagation Network (a form of a DBN) is initialized from a small amount of fully annotated data, and then refined by an EM-based learning method in an unsupervised fashion. During node refinement (the M step) a boosting-based algorithm is employed to train the evidence detectors of individual nodes. Experiments on a variety of data types - vision and inertial measurements - in several tasks demonstrate the ability to learn from as little as one fully annotated example accompanied by a small number of positive but non-annotated training examples. The system is applied to both recognition and anomaly detection tasks.},\n Doi = {10.1109/CVPR.2006.174},\n ISSN = {1063-6919},\n Keywords = {Application software;Boosting;Computer vision;Detectors;Graphical models;Hidden Markov models;Learning systems;Semisupervised learning;State-space methods;Surveillance},\n Timestamp = {2014.12.22}\n}\n\n","author_short":["Shi, Y.","Bobick, A.","Essa, I."],"key":"Shi2006","id":"Shi2006","bibbaseid":"shi-bobick-essa-learningtemporalsequencemodelfrompartiallylabeleddata-2006","role":"author","urls":{},"keyword":["Application software;Boosting;Computer vision;Detectors;Graphical models;Hidden Markov models;Learning systems;Semisupervised learning;State-space methods;Surveillance"],"downloads":0},"search_terms":["learning","temporal","sequence","model","partially","labeled","data","shi","bobick","essa"],"keywords":["application software;boosting;computer vision;detectors;graphical models;hidden markov models;learning systems;semisupervised learning;state-space methods;surveillance"],"authorIDs":[],"dataSources":["iCsmKnycRmHPxmhBd"]}