The results show from the left to the right: The original image sequence, the objects detected by the COST 211 Analysis Model (COST AM) [COST], the objects detected by an extended COST AM that considers moving cast shadows, and finally the detected and tracked moving cast shadows themselves.
![]() |
The results show, that shadow detection and tracking enhances markingly object segmentation. Cast shadows are detected and tracked separately. They are not any more detected as objects. Because the algorithm is based on an integration of observed signal changes, portions of the shadows that are covered by an object or portions that are always shadowed are not detected.
The following demos are available, too:
[an error occurred while processing this directive]