Show simple item record

dc.contributor.authorThangali, Ashwinen_US
dc.contributor.authorNash, Joanen_US
dc.contributor.authorSclaroff, Stanen_US
dc.contributor.authorNeidle, Carolen_US
dc.date.accessioned2015-06-24T19:48:33Z
dc.date.available2015-06-24T19:48:33Z
dc.date.issued2011-05-15
dc.identifier.citationThangali, Ashwin; Nash, Joan; Sclaroff, Stan; Neidle, Carol. "Exploiting Phonological Constraints for Handshape Inference in ASL Video", Technical Report BUCS-TR-2011-013, Computer Science Department, Boston University, May 15, 2011. [Available from: http://hdl.handle.net/2144/11370]
dc.identifier.urihttps://hdl.handle.net/2144/11370
dc.description.abstractHandshape is a key linguistic component of signs, and thus, handshape recognition is essential to algorithms for sign language recognition and retrieval. In this work, linguistic constraints on the relationship between start and end handshapes are leveraged to improve handshape recognition accuracy. A Bayesian network formulation is proposed for learning and exploiting these constraints, while taking into consideration inter-signer variations in the production of particular handshapes. A Variational Bayes formulation is employed for supervised learning of the model parameters. A non-rigid image alignment algorithm, which yields improved robustness to variability in handshape appearance, is proposed for computing image observation likelihoods in the model. The resulting handshape inference algorithm is evaluated using a dataset of 1500 lexical signs in American Sign Language (ASL), where each lexical sign is produced by three native ASL signers.en_US
dc.description.sponsorshipNational Science Foundation (0705749, 0855065)en_US
dc.language.isoen_US
dc.publisherComputer Science Department, Boston Universityen_US
dc.relation.ispartofseriesBUCS Technical Reports;BUCS-TR-2011-013
dc.titleExploiting phonological constraints for handshape inference in ASL videoen_US
dc.typeTechnical Reporten_US
dc.relation.isnodouble1152


This item appears in the following Collection(s)

Show simple item record