Reasoning About Object Appearance in The Context of a Scene

Dr. Mark Stevens
Computer Science Department
Colorado State University

Wednesday, March 10, 1999
11 a.m.
Fuller Labs 311

Automated recognition of objects is complicated by occlusion. Algorithms that operate reliably when objects are in plain view typically fail when more than 50 percent of one object is obscured by another. The reason, in short, is that these algorithms fail to find the majority of the object and therefore conclude it is not present. Moreover, since traditional techniques look for single objects in isolation, they cannot represent the fact that one object is obscuring the view of another. Instead, the occluded portions of objects must be labelled as not found. To explain the absence of the occluded features, a recognition system must use knowledge about an object's relationship to other objects in the scene. This talk discusses a way of recognizing occluded objects in the context of a scene. Partial scene models are constructed which explain absent features in terms of occlusion by other objects. The scene modeling is based upon minor enhancements to standard computer graphics techniques. Thus, a parameterized partial 3D scene model is used to generate predictions of how an image will appear. Recognition is then cast as a search for the parameters of a scene model which best matches the prediction to the observed imagery.

Host

Micha Hofri

Colloquium Coordinator

Carolina Ruiz

Maintained by webmaster@wpi.edu
Last modified: Sep 27, 2006, 16:05 EDT