Compound effects of top-down and bottom-up influences on visual attention during action recognition
File(s)
Accepted version
Author(s)
Khadhouri,B
Demiris,Y
Type
Conference Paper
Abstract
The limited visual and computational resources available during the perception of a human action makes a visual attention mechanism essential. In this paper we propose an attention mechanism that combines the saliency of top-down (or goal-directed) elements, based on multiple hypotheses about the demonstrated action, with the saliency of bottom-up (or stimulus-driven) components. Furthermore, we use the bottom-up part to initialise the top-down, hence resulting in a selection of the behaviours that rightly require the limited computational resources. This attention mechanism is then combined with an action understanding model and implemented on a robot, where we examine its performance during the observation of object-directed human actions.
Editor(s)
Kaelbling, LP
Saffiotti, A
Date Issued
2005-08
Citation
2005, pp.1458-1463
ISBN
9780938075936
0-9380-7593-4
Publisher
International Joint Conferences on Artificial Intelligence
Start Page
1458
End Page
1463
Journal / Book Title
19TH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI-05)
Copyright Statement
© 2005 International Joint Conference on Artificial Intelligence
Description
29.01.14 KB. Ok to add accepted version to spiral, publisher grants permission
Identifier
http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=000290233000231&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=1ba7043ffcc86c417c072aa74d649202
Source
International Joint Conference on Artificial Intelligence (IJCAI)
Source Place
Edinburgh, Scotland
Publication Status
Published
Start Date
2005-07-30
Finish Date
2005-08-05
Coverage Spatial
Edinburgh, Scotland