Abstract

A canonical problem in computer vision is category classification (e.g. find all instances of human faces, cars etc., in an image). Typically, the input for training a classifier is a relatively small sample of positive examples, and a much larger sample of negative examples, which in current applications can consist of images from thousands of categories. The difficulty of the problem sharply increases with the dimension and size of the negative example set. In this talk I will describe an efficient and easy to apply classification algorithm, which replaces the negative samples by a prior and then constructs a “hybrid” classifier which separates the positive samples from this prior. The resulting classifier achieves an identical or better classification rate than SVM , while requiring far smaller memory and lower computational complexity to train and apply. While here it is applied to image classes, the idea is general and can be applied to other domains. Joint work with Margarita Osadchy and Bella Fadida-Specktor. An early version of this work was presented in ECCV 2012 .