Show simple item record

dc.contributor.advisorTrevor Darrell
dc.contributor.authorMorency, Louis-Philippe
dc.contributor.authorQuattoni, Ariadna
dc.contributor.authorDarrell, Trevor
dc.contributor.otherVision
dc.date.accessioned2007-01-08T01:16:48Z
dc.date.available2007-01-08T01:16:48Z
dc.date.issued2007-01-07
dc.identifier.otherMIT-CSAIL-TR-2007-002
dc.identifier.urihttp://hdl.handle.net/1721.1/35276
dc.description.abstractMany problems in vision involve the prediction of a class label for each frame in an unsegmented sequence. In this paper we develop a discriminative framework for simultaneous sequence segmentation and labeling which can capture both intrinsic and extrinsic class dynamics. Our approach incorporates hidden state variables which model the sub-structure of a class sequence and learn the dynamics between class labels. Each class label has a disjoint set of associated hidden states, which enables efficient training and inference in our model. We evaluated our method on the task of recognizing human gestures from unsegmented video streams and performed experiments on three different datasets of head and eye gestures. Our results demonstrate that our model for visual gesture recognition outperform models based on Support Vector Machines, Hidden Markov Models, and Conditional Random Fields.
dc.format.extent8 p.
dc.format.extent366380 bytes
dc.format.extent946243 bytes
dc.format.mimetypeapplication/pdf
dc.format.mimetypeapplication/postscript
dc.language.isoen_US
dc.relation.ispartofseriesMassachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory
dc.titleLatent-Dynamic Discriminative Models for Continuous Gesture Recognition


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record