Abstract
The objective of this work is to identify and develop color and texture features for detection, analysis and matching of hair, working from non-frontal views (i.e. without the help of the presence of a face). As a secondary goal the student is expected to build a large database of hair pictures, taken from a back view of a cohort of human subjects.
Y. Yacoob and L. Davis. Detection, analysis and matching of hair. In Proc. Int'l. Conf. on Computer Vision, 2005.
From the same author:
http://www.umiacs.umd.edu/~yaser/pamis.pdf