We introduce a novel view-based object representation, called the saliency map graph (SMG), which captures the salient regions of an object view at multiple scales using a wavelet transform. This compact representation is highly invariant to translation, rotation (image and depth), and scaling, and offers the locality of representation required for occluded object recognition. To compare two saliency map graphs, we introduce two graph similarity algorithms. The first computes the topological similarity between two SMG's, providing a coarse-level matching of two graphs. The second computes the geometrical similarity between two SMG's, providing a fine-level matching of two graphs. We test and compare these two algorithms on a large database of model object views. Keywords: View-Based Object Recognition, Shape Representation and Recovery, Graph Matching. 1 Introduction The view-based approach to 3-D object recognition represents an object as a collection of 2-D views, sometimes called...

This paper is organized in the following way. Section 2 first provides a brief overview of the scheme for decomposing an image into the scale-space cells and computing the sMiency maps. The scheme is presented elsewhere in detail [5]. Section 3 deals with knowledge representation using Bayesian networks, whereas Section 4 deals with task-specific evidential reasoning and cooperative function of data-driven and knowledge-driven influences in selecting the areas of interest and gathering the evidence. The basis for the knowledge-driven selective perception and evidential reasoning part of the model presented here is Rimey and Brown's work [7, 8]. The implementation is described and the results are reported in Section 5. The implications of the results are discussed in Section 6