Publication Details
Motion Prediction Using Adaptive Association Rules
adaptive association rules, motion, prediction, time series, M4 project
This paper presents a method for object motion prediction in video. The method will work on XML data produced by M4 Project. The M4 project (Multi Modal Meeting Manager) is concerned with the construction of a demonstration system to enable structuring, browsing and querying of an archive of automatically analyzed meetings. Output XML data describe motions of some objects in video, but this data is not complete. So we need some prediction model which provides us information about position and movement of the objects in video sequences which are not described in output XML document. This paper is focused on a prediction method using adaptive association rules. These rules are generated during time series processing and they continuously adapt to current behavior of the time series representing the motion vector of an object.
@inproceedings{BUT22365,
author="Lukáš {Stryka}",
title="Motion Prediction Using Adaptive Association Rules",
booktitle="Proceedings of the 12th Conference STUDENT EEICT 2006 Volume 4",
year="2006",
pages="400--404",
publisher="Brno University of Technology",
address="Brno",
isbn="80-214-3163-6",
url="http://www.feec.vutbr.cz/EEICT/2006/sbornik/03-Doktorske_projekty/07-Informacni_systemy/11-stryka.pdf"
}