Recognition of Visual Dynamical Processes: Theory, Kernels, and Experimental Evaluation

Show simple item record

dc.contributor.author Vidal, René
dc.contributor.author Chaudhry, Rizwan
dc.date.accessioned 2009-04-14T18:52:25Z
dc.date.available 2009-04-14T18:52:25Z
dc.date.issued 2009
dc.identifier.uri http://jhir.library.jhu.edu/handle/1774.2/33295
dc.description.abstract Over the past few years, several papers have used Linear Dynamical Systems (LDS)s for modeling, registration, segmentation, and recognition of visual dynamical processes, such as human gaits, dynamic textures and lip articulations. The recognition framework involves identifying the parameters of the LDSs from features extracted from a training set of videos, using metrics on the space of dynamical systems to compare them, and combining these metrics with different classification methods. Usually, each paper makes an ad-hoc choice for every step, and tests the recognition framework on small data sets often involving only one application. We present a detailed evaluation of the LDS-based recognition pipeline; comparing identification methods, metrics, and classification techniques. We propose new metrics that have certain invariance properties and explore a number of variations to the existing metrics. We perform experimental evaluations on well-known data sets of human gaits, dynamic textures, and lip articulations and provide benchmark recognition results. We also analyze the robustness of the recognition pipeline with respect to changes in observation and experimental conditions. Overall, this work represents the most extensive to-date evaluation of the LDS-based recognition framework. en
dc.description.provenance Submitted by David Reynolds (davidr@jhu.edu) on 2009-04-14T15:35:15Z No. of bitstreams: 1 ChaudhryTR09.pdf: 1583326 bytes, checksum: ffc8759b69568214d3fd4f798ae6394c (MD5) en
dc.description.provenance Approved for entry into archive by Laura Graham(lgraham@jhu.edu) on 2009-04-14T18:52:25Z (GMT) No. of bitstreams: 1 ChaudhryTR09.pdf: 1583326 bytes, checksum: ffc8759b69568214d3fd4f798ae6394c (MD5) en
dc.description.provenance Made available in DSpace on 2009-04-14T18:52:25Z (GMT). No. of bitstreams: 1 ChaudhryTR09.pdf: 1583326 bytes, checksum: ffc8759b69568214d3fd4f798ae6394c (MD5) Previous issue date: 2009 en
dc.description.sponsorship This work was partially supported by startup funds from JHU, by grants ONR N00014-05-10836, NSF CAREER 0447739, NSF EHS-0509101, and by contract JHU APL-934652. en
dc.language.iso en_US en
dc.relation.ispartofseries Department of Computer Science, April 2009;Technial Report 09-01
dc.subject Classification en
dc.subject Kernels for time series data en
dc.subject Linear dynamical systems en
dc.subject Dynamic textures en
dc.subject Action recognition en
dc.title Recognition of Visual Dynamical Processes: Theory, Kernels, and Experimental Evaluation en
dc.type Working Paper en

Files in this item

Files Size Format Download
ChaudhryTR09.pdf 1.583Mb application/pdf Download

This item appears in the following Collection(s)

Show simple item record