Patent attributes
A video encoder and corresponding method are provided for encoding video signal data for an image block and a particular reference picture index to predict the image block, where the encoder includes a fast search block motion estimator for providing motion vectors corresponding to the at least one particular reference picture, the motion estimator comprising a fast search block matching portion for performing fast search block matching while excluding non-optimal search points in accordance with a comparison of a normalization of the image block pixels against a normalization of the reference picture pixels, the fast search block matching portion having an output responsive to the at least one particular reference picture; and the corresponding method includes receiving a substantially uncompressed image block, block matching the image block in correspondence with at least one particular reference picture while excluding non-optimal search points in accordance with a comparison of a normalization of the image block pixels against a normalization of the reference picture pixels, computing motion vectors corresponding to a difference between the image block and the at least one particular reference picture, and motion compensating the at least one particular reference picture in correspondence with the motion vectors.
