CSAIL Event Calendar: Previous Series
|
Object Detection Using Semi-NaÔve Bayes to Model Sparse Structure Speaker: Henry Schneiderman , Carnegie Mellon University Many classes of images have sparse structuring of statistical dependency. Each variable has strong statistical dependency with a small number of other variables and negligible dependency with the remaining ones. Sparse structure simplifies the task of recognizing various objects. In particular, a semi-naÔve Bayes classifier compactly represents sparseness. A semi-naÔve Bayes classifier decomposes the input variables into subsets and represents statistical dependency within each subset, while treating the subsets as statistically independent. This talk describes an automatic method for constructing a semi-naÔve Bayes classifier for object detection. This method generates a pool of candidate subsets where each subset captures a significant statistical dependency. The method then trains a "sub-classifier" over each such subset. Empirical techniques select a group of these sub-classifiers to form the final classifier. This approach achieves reliable and efficient detection for several objects including faces, eyes, ears, telephones, push-carts, and door-handles.
See other events that are part of Brains & Machines Seminar Series Fall 2002 |
