Show simple item record

dc.contributor.authorLu, Yao
dc.contributor.supervisorProf. Ling Li
dc.contributor.supervisorDr Patrick Peursum
dc.date.accessioned2017-01-30T10:10:59Z
dc.date.available2017-01-30T10:10:59Z
dc.date.created2014-01-17T05:32:21Z
dc.date.issued2013
dc.identifier.urihttp://hdl.handle.net/20.500.11937/1665
dc.description.abstract

This thesis describes a bottom-up approach to estimating human pose over time based on monocular views with no restriction on human activities,Three approaches are proposed to address the weaknesses of existing approaches, including building a specific appearance model using clustering,utilising both the generic and specific appearance models in the estimation, and building an uncontaminated appearance model by removing backgroundpixels from the training samples. Experimental results show that the proposed system outperforms existing system significantly.

dc.languageen
dc.publisherCurtin University
dc.titleHuman body tracking and pose estimation from monocular image sequences
dc.typeThesis
dcterms.educationLevelPh.D.
curtin.departmentSchool of Electrical Engineering and Computing, Department of Computing
curtin.accessStatusOpen access


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record