Object tracking by chiaroscuro patterns

Chiaroscuro in art is characterized by strong contrasts between light and dark. An object in a certain light condition has a certain chiaroscuro pattern in appearance; and this pattern is invariant to the changes of illumination and pose within a certain extent. In this paper we introduce an object tracking method based on chiaroscuro patterns. The chiaroscuro pattern of an image is computed by running a medial axis transform on selected level sets. The medial axes are pruned and decomposed into segments. These segments make a chiaroscuro pattern. The end points of the segments are collected as key points, which are fed to Kalman filter for object tracking. The effectiveness of the algorithm is demonstrated by experiments.