Model driven edge detection

Standard edge detectors fail to find most relevant edges, finding either too many or too few, because they lack a geometric model to guide their search. We present a technique that integrates both photometric and geometric models with an initial estimate of the boundary. The strength of this approach lies in the ability of the geometric model to overcome various photometric anomalies, thereby finding boundaries that could not otherwise be found. Furthermore, edges can be scored based on their goodness of fit to the model, thus allowing one to use semantic model information to accept or reject the edges.