Scene modelling and classification using learned spatial relations

H...............H

Show simple item record

dc.contributor.author Hogg, David C.
dc.contributor.author Cohn, Anthony G.
dc.contributor.author Dee, Hannah
dc.date.accessioned 2010-11-15T14:08:58Z
dc.date.available 2010-11-15T14:08:58Z
dc.date.issued 2009
dc.identifier.citation Hogg , D C , Cohn , A G & Dee , H 2009 , ' Scene modelling and classification using learned spatial relations ' pp. 295-311 . en
dc.identifier.other PURE: 152837
dc.identifier.other dspace: 2160/5874
dc.identifier.uri http://hdl.handle.net/2160/5874
dc.description Dee, H. M., Hogg, D. C., and Cohn, A. G. Scene Modelling and Classification Using Learned Spatial Relations Springer LNCS (Conference on Spatial Information Theory (COSIT)) pp 295-311, l'Aber Wrac'h, France, September 2009. Sponsorship: EPSRC en
dc.description.abstract This paper describes a method for building visual scene models from video data using quantized descriptions of motion. This method enables us to make meaningful statements about video scenes as a whole (such as “this video is like that video”) and about regions within these scenes (such as “this part of this scene is similar to this part of that scene”). We do this through unsupervised clustering of simple yet novel motion descriptors, which provide a quantized representation of gross motion within scene regions. Using these we can characterise the dominant patterns of motion, and then group spatial regions based upon both proximity and local motion similarity to define areas or regions with particular motion characteristics. We are able to process scenes in which objects are difficult to detect and track due to variable frame-rate, video quality or occlusion, and we are able to identify regions which differ by usage but which do not differ by appearance (such as frequently used paths across open space). We demonstrate our method on 50 videos making up very different scene types: indoor scenarios with unpredictable unconstrained motion, junction scenes, road and path scenes, and open squares or plazas. We show that these scenes can be clustered using our representation, and that the incorporation of learned spatial relations into the representation enables us to cluster more effectively. en
dc.format.extent 17 en
dc.language.iso eng
dc.relation.ispartof en
dc.title Scene modelling and classification using learned spatial relations en
dc.type Text en
dc.type.publicationtype Conference paper en
dc.identifier.doi http://dx.doi.org/10.1007/978-3-642-03832-7_18
dc.contributor.institution Department of Computer Science en
dc.contributor.institution Vision, Graphics and Visualisation Group en
dc.description.status Non peer reviewed en


Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record

Search Cadair


Advanced Search

Browse

My Account