J.H. von Brecht, S.R. Thiruvenkadam, and T.F. Chan (USA)
Image segmentation, tracking, variational methods
We present a variational PDE based model for tracking ob jects under occlusion. Here, prior shape information is used to segment object boundaries that are occluded. The novelty in this work is that the shape prior is combined with the image term using logical operations pertaining to a unique occlusion scenario, thus avoiding locally optimal solutions. The model was tested on real and synthetic im age sequences with promising results.
Important Links:
Go Back
