X-Y separable pyramid steerable scalable kernels

A new method for generating x-y separable steerable scalable approximations offilter kemels is proposed which is based on a generalization of the Singular Value Decomposition (SVD) to 3 dimensions. This “pseudo-SVD ” impmves upon a p w w u s scheme due to Pemna in that it reduces convolution time and storage requirements. An adaptation of the pseudo-SVD is proposed to generate steerable and scalable kemels which an? suitable for use with a Laplacian pyramid. The properties of this method are illustrated experimentally in generating steerable and scalable approximations to an early vision edge-detection bmel. keywords: steerable, scalable filters, multi-msohtion multi-orientation filtering, early vision, multi-way arrays, S W , pyramid