Show simple item record

dc.contributor.advisorTomaso Poggio
dc.contributor.authorNi, Yuzhaoen_US
dc.contributor.authorFrogner, Charles A.en_US
dc.contributor.authorPoggio, Tomaso A.en_US
dc.contributor.otherCenter for Biological and Computational Learning (CBCL)en_US
dc.date.accessioned2013-09-19T22:30:06Z
dc.date.available2013-09-19T22:30:06Z
dc.date.issued2013-09-19
dc.identifier.urihttp://hdl.handle.net/1721.1/80815
dc.description.abstractIn this thesis, we designed and implemented a crowdsourcing system to annotatemouse behaviors in videos; this involves the development of a novel clip-based video labeling tools, that is more efficient than traditional labeling tools in crowdsourcing platform, as well as the design of probabilistic inference algorithms that predict the true labels and the workers' expertise from multiple workers' responses. Our algorithms are shown to perform better than majority vote heuristic. We also carried out extensive experiments to determine the effectiveness of our labeling tool, inference algorithms and the overall system.en_US
dc.format.extent69 p.en_US
dc.relation.ispartofseriesMIT-CSAIL-TR-2013-023
dc.relation.ispartofseriesCBCL-314
dc.subjectcrowdsourcingen_US
dc.subjectvideo labelingen_US
dc.subjecthuman computationen_US
dc.subjectmouse phenotypingen_US
dc.subjectaction recognitionen_US
dc.titleMouse Behavior Recognition with The Wisdom of Crowden_US
dc.date.updated2013-09-19T22:30:06Z


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record