Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
10359170 | Computer Vision and Image Understanding | 2013 | 14 Pages |
Abstract
⺠We introduce a novel method to fuse arbitrary single target tracking approaches. ⺠We show how to transform diverse tracking outputs to a common representation. ⺠Object segmentations are iteratively computed from the fused tracking outputs. ⺠Performance based weighting of the trackers further increases the accuracy. ⺠We fused three complementary tracking approaches in our experiments.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Vision and Pattern Recognition
Authors
Markus Heber, Martin Godec, Matthias Rüther, Peter M. Roth, Horst Bischof,