Publications

2006

N. Dowson and R. Bowden
N-tier Simultaneous Modelling and Tracking for Arbitrary Warps.
In: Proceedings of the 17th British Machine Vision Conference, Edinburgh, UK, September 2006.
Download: pdf 611k bib
Test Scripts: tgz 257k. (Try the trackers in "downloads" if you'd prefer the algorithms rather than test scripts.)
Demos:
Tracking amorphous objects despite difficult lighting conditions for a Hand and Person in a lobby.
Comparison of Tracking using Sum of Square Differences and Mutual Information.
Comparison of Tracking using the Strategic update model and Simultaneous Modelling and Tracking (SMAT).
Comparison of tracking with No, Thin and Thick structure models.

N. Dowson, R. Bowden and T. Kadir.
Image template matching using Mutual Information and NP-Windows.
In: Proceedings of the 18th International Conference on Pattern Recognition, Hong Kong, August 2006.
Download: pdf 410k bib
Test Scripts: warthogTst.tgz 3.4M

N. Dowson and R. Bowden.
A Unifying Framework for Mutual Information methods for use in Non-linear Optimisation.
In: Proceeding of the 9th European Conference on Computer Vision , Graz, Austria, May 2006.
Download: pdf 447k bib
Test Scripts: tgz 1316k (Try the warthog module in "downloads" if you'd prefer the algorithms rather than test scripts.)
Note that these test scripts have been obsoleted by the unified test script: warthogTst.tgz 3.4M

2005

N. Dowson and R. Bowden.
Simultaneous modeling and tracking (SMAT) of feature feature sets.
In: Proceeding of the IEEE Computer Society International Conference on Computer Vision and Pattern Recognition , San Diego, USA, June 2005.
Download: pdf 342k bib
Demos:
Comparing Tracking with and without a structure model for a signing lady (347k) and a rhino (264k).
Comparing Tracking with pre and post-learned a structure model for signing lady (334k) and a rhino (258k).

2004

N. Dowson and R. Bowden.
Metric Mixtures for Mutual Information Tracking.
In: Proceedings of the 17th International Conference on Pattern Recognition, Cambridge UK, August 2004.
Download: pdf 132k bib
Demos:
Comparing various update methods (none, naive, strategic and SMAT) for a newscaster (135k).
Example of tracking the torso of a pedestrian (265k) using SMAT.

Note