Pictorial Structures for Object Recognition：（对象识别的图形结构）.pdfVIP

Pictorial Structures for Object Recognition Pedro F. Felzenszwalb Artificial Intelligence Lab, Massachusetts Institute of Technology pff@ai.mit.edu Daniel P. Huttenlocher Computer Science Department, Cornell University dph@cs.cornell.edu Abstract In this paper we present a computationally efficient framework for part-based mod- eling and recognition of objects. Our work is motivated by the pictorial structure models introduced by Fischler and Elschlager. The basic idea is to represent an ob- ject by a collection of parts arranged in a deformable configuration. The appearance of each part is modeled separately, and the deformable configuration is represented by spring-like connections between pairs of parts. These models allow for qualitative descriptions of visual appearance, and are suitable for generic recognition problems. We address the problem of finding instances of an object in an image using pictorial structure models as well as the problem of learning an object model from training ex- amples, presenting efficient algorithms in both cases. We demonstrate the techniques by learning models that represent faces and human bodies and using the resulting models to locate the corresponding objects in novel images. 1 Introduction Research in object recognition is increasingly concerned with the ability to recognize generic classes of objects rather than just specific instances. In this paper, we con- sider both the problem of recognizing objects using generic part-based models and the problem of learning such models from example images. Our work is motivated by the pictorial structure representation introduced by Fischler and Elschlager [16] 1 (a) (b) Figure 1: Sample results for detection