Object recognition using steerable filters at multiple scales

Abstract

The identification and location of objects in images is difficult owing to the view variance of geometric features. This problem can be solved by developing view-insensitive descriptions of image points. View-insensitive descriptions are achieved by describing points in terms of the responses of steerable filters at multiple scales. The steerability allows the normalizing of rotations about the view vector. Owing to the use of multiple scales, the vector for each point is, for all practical purposes, unique, and thus can be easily matched to other instances of the point in other images.<>