An object recognition system has been developed that uses a new class of local image features. The features are invariant to image scaling, translation, and rotation, and partially invariant to illumination changes and affine or 3D projection. These features share similar properties with neurons in inferior temporal cortex that are used for object recognition in primate vision. Features are efficiently detected through a staged filtering approach that identifies stable points in scale space. Image keys are created that allow for local geometric deformations by representing blurred image gradients in multiple orientation planes and at multiple scales. The keys are used as input to a nearest-neighbor indexing method that identifies candidate object matches. Final verification of each match is achieved by finding a low-residual least-squares solution for the unknown model parameters. Experimental results show that robust object recognition can be achieved in cluttered partially-occluded images with a computation time of under 2 seconds.

In this paper we propose a new face recognition system based on a biologically-inspired filtering method. Our work differs from previous proposals in (1) the multi-stage filtering method employed, in (2) the pyramid structure used, and most importantly, in (3) the prototype construction scheme to determine the models stored in memory. The method is much simpler than previous proposals and relatively inexpensive computationally, while attaining error rates as low as 5%, very close to the best reported results. I. Introduction Automatic face recognition is an extremely important task. Boosted by commercial demands for face recognition systems, a wide range of proposals has been advanced in the present decade. Starting with the influential Eigenfaces approach by Turk and Pentland [10], a number of approaches have attained high levels of performance in face databases, in many cases with recognition rates over 90%. Besides the successful Eigenfaces approach, researchers have proposed...