Efficient Object-Class Recognition by Boosting Contextual Information
Object-class recognition is one of the most challenging fields of pattern recognition and computer vision. Currently, most authors represent an object as a collection of parts and their mutual spatial relations. Therefore, two types of information are extracted: local information describing each part, and contextual information describing the (spatial) context of the part, i.e. the spatial relations between the rest of the parts and the current one. We define a generalized correlogram descriptor and represent the object as a constellation of such generalized correlograms. Using this representation, both local and contextual information are gathered into the same feature space. We take advantage of this representation in the learning stage, by using a feature selection with boosting that learns both types of information simultaneously and very efficiently. Simultaneously learning both types of information proves to be a faster approach than dealing with them separately. Our method is compared with state-of-the-art object-class recognition systems by evaluating both the accuracy and the cost of the methods.
Unable to display preview. Download preview PDF.
- 1.Agarwal, S., Awan, A., Roth, D.: Learning to detect objects in images via a sparse, part-based representation. IEEE TPAMI 26, 1475–1490 (2004)Google Scholar
- 2.Schneiderman, H.: Learning a restricted bayesian network for object detection. In: IEEE Proc. CVPR, pp. 639–646 (2004)Google Scholar
- 3.R., F., P., P., A., Z.: Object class recognition by unsupervised scale-invariant learning. In: IEEE Proc. CVPR (2003) Google Scholar
- 5.Weber, M., Welling, M., Perona, P.: Towards automatic discovery of object categories. In: IEEE Proc. CVPR, pp. 101–108 (2000)Google Scholar
- 6.Huang, J., Kumar, S., Mitra, M., Zhu, W., Zabih, R.: Image indexing using color correlograms. In: IEEE Proc. CVPR, pp. 762–768 (1997)Google Scholar
- 7.Belongie, S., Malik, J., Puzicha, J.: Shape matching and object recognition using shape contexts. IEEE Trans. PAMI 24, 509–522 (2002)Google Scholar
- 10.Fei-Fei, L., Fergus, R., Perona, P.: A bayesian approach to unsupervised one-shot learning of object categories. In: IEEE Proc. ICCV, vol. 2, pp. 1134–1142 (2003)Google Scholar
- 11.Thayananthan, A., Stenger, B., Torr, P., Cipolla, R.: Shape context and chamfer matching in cluttered scenes. In: IEEE Proc. CVPR (2003)Google Scholar
- 12.Squire, M.D., Muller, H., Muller, W.: Improving response time by search pruning in a content-based image retrieval system, using inverted file techniques. In: IEEE Workshop CBAIVL (1999)Google Scholar