Weak Classifier for Density Estimation in Eye Localization and Tracking.(July 2017)
The main stray of this project is a fast weak classifier that can detect and track eyes in video sequences.
We have presented the IPD, a weak classifier that finds the probability distribution functions of desired features (in this paper, eyes) in images. It is simple and fast, and it naturally fits well in a Bayesian framework, such as the Kalman or particle filter, for tracking features in video sequences.