dc.contributor.author | Shakhnarovich, Gregory | en_US |
dc.contributor.author | Viola, Paul | en_US |
dc.contributor.author | Darrell, Trevor | en_US |
dc.date.accessioned | 2004-10-08T20:38:53Z | |
dc.date.available | 2004-10-08T20:38:53Z | |
dc.date.issued | 2003-04-18 | en_US |
dc.identifier.other | AIM-2003-009 | en_US |
dc.identifier.uri | http://hdl.handle.net/1721.1/6715 | |
dc.description.abstract | Example-based methods are effective for parameter estimation problems when the underlying system is simple or the dimensionality of the input is low. For complex and high-dimensional problems such as pose estimation, the number of required examples and the computational complexity rapidly becme prohibitively high. We introduce a new algorithm that learns a set of hashing functions that efficiently index examples relevant to a particular estimation task. Our algorithm extends a recently developed method for locality-sensitive hashing, which finds approximate neighbors in time sublinear in the number of examples. This method depends critically on the choice of hash functions; we show how to find the set of hash functions that are optimally relevant to a particular estimation problem. Experiments demonstrate that the resulting algorithm, which we call Parameter-Sensitive Hashing, can rapidly and accurately estimate the articulated pose of human figures from a large database of example images. | en_US |
dc.format.extent | 12 p. | en_US |
dc.format.extent | 5030222 bytes | |
dc.format.extent | 6836715 bytes | |
dc.format.mimetype | application/postscript | |
dc.format.mimetype | application/pdf | |
dc.language.iso | en_US | |
dc.relation.ispartofseries | AIM-2003-009 | en_US |
dc.subject | AI | en_US |
dc.subject | parameter estimation | en_US |
dc.subject | nearest neighbor | en_US |
dc.subject | locally weighted learning | en_US |
dc.title | Fast Pose Estimation with Parameter Sensitive Hashing | en_US |