Show simple item record  

dc.contributor.authorZhang, Edmond Yiwen
dc.contributor.authorMayo, Michael
dc.coverage.spatialConference held at Christchurch, New Zealanden_NZ
dc.date.accessioned2009-05-21T01:54:36Z
dc.date.available2009-05-21T01:54:36Z
dc.date.issued2008
dc.identifier.citationZhang, E. & Mayo, M. (2008). Pattern discovery for object categorization. In Proceeding of 23rd International Conference Image and Vision Computing New Zealand 2008(IVCNZ 2008).en
dc.identifier.urihttps://hdl.handle.net/10289/2173
dc.description.abstractThis paper presents a new approach for the object categorization problem. Our model is based on the successful `bag of words' approach. However, unlike the original model, image features (keypoints) are not seen as independent and orderless. Instead, our model attempts to discover intermediate representations for each object class. This approach works by partitioning the image into smaller regions then computing the spatial relationships between all of the informative image keypoints in the region. The results show that the inclusion of spatial relationships leads to a measurable increase in performance for two of the most challenging datasets.en
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.publisherIEEE Pressen_NZ
dc.relation.urihttp://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=4762071en
dc.rightsThis article has been published in the Proceeding of 23rd International Conference Image and Vision Computing New Zealand 2008 (IVCNZ 2008). ©2008 IEEE.en
dc.subjectcomputer scienceen
dc.subjectimage processingen
dc.subjectkeypointsen
dc.subjectrecognitionen
dc.subjectcategorizationen
dc.subjectMachine learning
dc.titlePattern discovery for object categorizationen
dc.typeConference Contributionen
dc.identifier.doi10.1109/IVCNZ.2008.4762071en
dc.relation.isPartOfImage and Vision Computing New Zealand, 23rd International Conferenceen_NZ
pubs.begin-page1en_NZ
pubs.elements-id18469
pubs.end-page6en_NZ
pubs.finish-date2008-11-28en_NZ
pubs.start-date2008-11-26en_NZ


Files in this item

This item appears in the following Collection(s)

Show simple item record