Invariant features for gray scale images

Abstract

Invariant features are image characteristics which remain unchanged under the action of a transformation group. We consider in this paper image rotations and translations and present algorithms for constructing invariant features. After briefly sketching the theoretical background we develop algorithms for recognizing several objects in a single scene without the necessity to segment the image beforehand. The objects can be rotated and translated independently. Moderate occlusions are tolerable. Furthermore we show how to use these techniques for the recognition of articulated objects. The methods work directly with the gray values and do not rely on the extraction of geometric primitives like edges or corners in a preprocessing step. All algorithms have been implemented and tested both on synthetic and real image data. We present some illustrative experimental results.