This paper proposes new hypergeometric" lters for the problem of image matching under the translational and a ne model. This new set of lters has the following advantages: 1 High-precision registration of two images under the translational and a ne model. Because the window e ects are eliminated, we are able to achieve superb performance i n b oth translational and a ne matching. 2 A ne matching without exhaustive search or image warping. Due to the recursiveness of the lters in the spatial

more »

... in, we are able to analytically express the relation between lter outputs and the six a ne parameters. This analytical relation enables us to directly compute these a ne parameters. 3 Generality. The approach we demonstrate here can be applied t o a b r oad class of matching problems, as long as the transformation between the two image patches can be mathematically represented in the frequency domain.