Weak Models and Cue Integration for Real-Time Tracking
2002 (English)Conference paper (Refereed)
Traditionally, fusion of visual information for tracking has been based on explicit models for uncertainty and integration. Most of the approaches use some form of Bayesian statistics where strong models are employed. We argue that for cases where a large number of visual features are available, weak models for integration may be employed. We analyze integration by voting where two methods are proposed and evaluated: i) response and ii) action fusion. The methods differ in the choice of Voting space: the former integrates visual information in image space and latter in velocity space. We also evaluate four weighting techniques for integration.
Place, publisher, year, edition, pages
2002. 3044-3049 p.
Computer and Information Science
IdentifiersURN: urn:nbn:se:kth:diva-82181OAI: oai:DiVA.org:kth-82181DiVA: diva2:498001
19th IEEE International Conference on Robotics and Automation
NR 201408052012-02-112012-02-11Bibliographically approved