Research/Color-based visual model with background information
When tracking nonrigid objects like persons’ bodies or hands, color histograms, sampled from elliptical regions, provide a convenient way to encode the visual properties of these objects.
However, when the color of the background is similar to the color of the tracked object, the classical approach of color histograms will fail. This is illustrated below with an image of a white person on a white background. Note that the measure of presence is severely noisy and multi-modal, therefore the person’s position cannot be easily determined.
To improve detection in situations like depicted above, we have modified the measure of presence through use of an estimate of the background image and the normalized Hellinger distance. This measure is depicted in the image below.
An example of tracking on a cluttered background with a standard histogram-based measure and the proposed visual model is shown below.
Kristan, M. Tracking people in video data using probabilistic models Ph.D. Thesis, Faculty of Electrical Engineering, May 2008.
Kristan, M. and Perš, J. and Perše, M. and Kovačič S. “Closed-world tracking of multiple interacting targets for indoor-sports applications”. Computer Vision and Image Understanding, 2008.