US 6,320,906 B1 · Motion vector detecting circuit

Abstract

In an animation encoding process, a motion vector which can minimize an AC component included in a discrete cosine transform result is detected. For this purpose, a block X (i, j) to be processed is extracted from an input image X, and a prospective reference block Yk (i, J) is extracted from a reference image Y, wherein k is an integer depending upon the size of a motion vector search range, i=0, . . . , or 3, and j=0, . . . , or 63. From each pixel value in each sub block (including 8×8 pixels) of the block to be processed, an average value Xa (i) of pixels in the sub block is subtracted, and from each pixel value in each sub block (including 8×8 pixels) of the prospective reference block, an average value Yka (i) of pixels in the sub block is subtracted. Then, an evaluation value Sk for block matching is calculated. The prospective reference block having the minimum evaluation value is selected as a reference block among the plural prospective reference blocks, so as to obtain a motion vector V corresponding to the positional relationship between the block to be processed and the reference block.

A motion vector detecting circuit for detecting a motion vector in an input image and a reference image, said circuit comprising:

means for extracting, from said input image, a block to be processed which includes plural pixels;
means for dividing said block to be processed into plural sub blocks, each of said sub blocks having the same size as a processing unit in discrete cosine transform (DCT);
means for extracting, from a given search range in said reference image, plural prospective reference blocks each of which includes pixels in the same number as that of said pixels included in said block to be processed;
means for dividing each of said plural prospective reference blocks into plural sub blocks in the same number as that of said sub blocks of said block to be processed, each of said sub blocks of said plural prospective reference blocks having the same size as that of said sub blocks of said block to be processed;
calculating means for calculating, with regard to each of said plural prospective reference blocks, an evaluation value corresponding to similarity between said block to be processed and said prospective reference block by using a result obtained by subtracting an average value of pixels included in each of said sub blocks of said block to be processed from respective pixel values of each of said sub blocks of said block to be processed and by using a result obtained by subtracting an average value of pixels included in each of said sub blocks of said prospective reference block from respective pixel values of each of said sub blocks of said prospective reference block;
means for comparing said evaluation values of said plural prospective reference blocks in order to determine which of said plural prospective reference blocks is most similar to said block to be processed;
and
means for generating said motion vector on the basis of a result of comparison,
wherein an image difference between each of sub blocks of said input image and each of sub blocks of said reference image is obtained by using said motion vector, and the DCT is carried out to the image difference so as to effect a reduction of an AC component included in the DCT result;
and
wherein said calculating means includes:
means for calculating an average value of pixels included in each of said sub blocks of said block to be processed;
means for obtaining a first intermediate block including pixels resulting from subtracting said average value of said pixels included in each of said sub blocks from respective pixel values of each of said sub blocks of said block to be processed;
means for calculating an average value of pixels included in each of said sub blocks of said prospective reference block;
means for obtaining a second intermediate block including pixels resulting from subtracting said average value of said pixels included in each of said sub blocks of said prospective reference block from respective pixel values of each of said sub blocks of said prospective reference block;
means for obtaining a third intermediate block including pixels resulting from subtracting corresponding pixel values in said second intermediate block from pixel values in said first intermediate block;
and
means for calculating said evaluation value corresponding to the similarity between said block to be processed and said prospective reference block by using respective pixel values of said third intermediate block.

15 more elements

Motion vector detecting circuit

Life of this patent

Motion vector detecting circuit

Life of this patent

Abstract

Claims5