Output details
11 - Computer Science and Informatics
University of York
Article title
Coupled Prediction Classification for Robust Visual Tracking
Type
D - Journal article
Title of journal
IEEE Transactions on Pattern Analysis and Machine Intelligence
Article number
5288528
Volume number
32
Issue number
9
First page of article
1553
ISSN of journal
0162-8828
Year of publication
2010
Number of additional authors
1
Additional information
<23>Tracking is theoretically formulated as a Bayesian mixture of experts, and this model is used for regression and hence data relevance determination. Experimentally, the method is extensively evaluated on a large number of standard motion data-sets posing problems of varying difficulty, and compared to the state of the art alternatives. The regression-classification method proposed may prove to be a general use in machine vision, beyond the tracking problem studied in the paper. Resulted in successful DSTL CDE proposal on face tracking from partial views in poor quality CCTV fragments.
Interdisciplinary
-
Cross-referral requested
-
Research group
H - Computer Vision and Pattern Recognition
Citation count
5
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
