Cascade object detection with deformable part models

We describe a general method for building cascade classifiers from part-based deformable models such as pictorial structures. We focus primarily on the case of star-structured models and show how a simple algorithm based on partial hypothesis pruning can speed up object detection by more than one order of magnitude without sacrificing detection accuracy. In our algorithm, partial hypotheses are pruned with a sequence of thresholds. In analogy to probably approximately correct (PAC) learning, we introduce the notion of probably approximately admissible (PAA) thresholds. Such thresholds provide theoretical guarantees on the performance of the cascade method and can be computed from a small sample of positive examples. Finally, we outline a cascade detection algorithm for a general class of models defined by a grammar formalism. This class includes not only tree-structured pictorial structures but also richer models that can represent each part recursively as a mixture of other parts.

1. Introduction

A popular approach for object detection involves reducing the problem to binary classification. The simplest and most common example of this approach is the sliding window method. In this method a classifier is applied at all positions, scales, and, in some cases, orientations of an image. However, testing all points in the search space with a non-trivial classifier can be very slow. An effective method for addressing this problem involves applying a cascade of simple tests to each hypothesized object location to eliminate most of them very quickly [16], [12], [4], [15], [2], [13].