Abstract

The ma jority of current methods in ob ject classification use the one-against-rest training scheme. We argue that when applied to a large number of classes, this strategy is problematic: as the number of classes increases, the negative class becomes a very large and complicated collection of images. The resulting classification problem then becomes extremely unbalanced, and kernel SVM classifiers trained on such sets require long training time and are slow in prediction. To address these problems, we propose to consider the negative class as a background and characterize it by a prior distribution. Further, we propose to construct ”hybrid” classifiers, which are trained to separate this distribution from the samples of the positive class. A typical classifier first pro jects (by a function which may be non-linear) the inputs to a one-dimensional space, and then thresholds this pro jection. Theoretical results and em- pirical evaluation suggest that, after pro jection, the background has a relatively simple distribution, which is much easier to parameterize and work with. Our results show that hybrid classifiers offer an advantage over SVM classifiers, both in performance and complexity, especially when the negative (background) class is large.