Relevant Feature Selection for Human Pose Estimation and Localization in Cluttered Images Ryuzo Okada and Stefano Soatto Computer Science Department, University of California, Los Angeles {okada,soatto - Okada - Document - PDFSEARCH.IO

First Page	Meta Content
	Relevant Feature Selection for Human Pose Estimation and Localization in Cluttered Images Ryuzo Okada and Stefano Soatto Computer Science Department, University of California, Los Angeles {okada,soatto}@cs.ucla.edu Add to Reading List Document Date: 2008-08-28 23:54:49 Open Document File Size: 1,29 MB Share Result on Facebook Company Toshiba Corporation / Intel / / / Facility University of California / / IndustryTerm post-processing / type algorithms / human model rendering software / sparse solution / inner product / sparse solutions / wide search range / / Organization University of California / Los Angeles / Cluttered Images Ryuzo Okada and Stefano Soatto Computer Science Department / / Position driver / / Product HumanEva-I / / ProgrammingLanguage R / / ProvinceOrState California / / RadioStation KRVM / / Technology Expectation-Maximization type algorithms / quasi-Newton algorithm / machine learning / 3-D / / URL http / SocialTag Robot control Regression analysis Statistical classification Support vector machine Linear regression Pose K-nearest neighbor algorithm 3D pose estimation Non-negative matrix factorization Statistics

Relevant Feature Selection for Human Pose Estimation and Localization in Cluttered Images Ryuzo Okada and Stefano Soatto Computer Science Department, University of California, Los Angeles {okada,soatto}@cs.ucla.eduAdd to Reading List