Date Added: Sep 2009
The authors introduce Active Hidden Models (AHM) that utilize kernel methods traditionally associated with classification. They use AHMs to track deformable objects in video sequences by leveraging kernel projections. They introduce the "Subset projection" method which improves the efficiency of the tracking approach by a factor of ten. They successfully tested the method on facial tracking with extreme head movements (including full 180-degree head rotation), facial expressions, and deformable objects. Given a kernel and a set of training observations, they derive unbiased estimates of the accuracy of the AHM tracker. Kernels are generally used in classification methods to make training data linearly separable.