Method for estimating motion in a video sequence

Claim 1

A method for estimating motion within a video sequence, the method comprising the steps of: a) performing a first series of comparisons between a present video frame and a comparison video block of a previously stored video frame, wherein, for each comparison of the first series of comparisons, the comparison video block is incrementally positioned within a first region of the previously stored video frame based on a first region of the previously stored video frame based on a first search point displacement pattern; b) performing a second series of comparisons between the present video block and the comparison video block, wherein, for each comparison of the second series of comparisons, the comparison video block is incrementally positioned within a second region of the previously stored video frame based on a second search point displacement pattern, and wherein the second search point displacement pattern is greater than the first search point displacement pattern; c) performing a third series of comparisons between the present video block and the comparison video block, wherein, for each comparison of the third series of comparisons, the comparison video block is incrementally positioned within a third region of the previously stored video frame based on a third region of the previously stored video frame based on a third search point displacement pattern is greater than the second search point displacement pattern; and d) from the first series of comparisons, the second series of comparisons and the third series of comparisons, determining a match video block, further comprising establishing the first region, the second region, and the third region of the previously stored video frame to have a rectangular geometry.

Claim 2

A method for estimating motion within a video sequence, the method comprising the steps of: a) performing a first series of comparisons between a present video frame and a comparison video block of a previously stored video frame, wherein, for each comparison of the first series of comparisons, the comparison video block is incrementally positioned within a first region of the previously stored video frame based on a first region of the previously stored video frame based on a first search point displacement pattern; b) performing a second series of comparisons between the present video block and the comparison video block, wherein, for each comparison of the second series of comparisons, the comparison video block is incrementally positioned within a second region of the previously stored video frame based on a second search point displacement pattern, and wherein the second search point displacement pattern is greater than the first search point displacement pattern; c) performing a third series of comparisons between the present video block and the comparison video block, wherein, for each comparison of the third series of comparisons, the comparison video block is incrementally positioned within a third region of the previously stored video frame based on a third region of the previously stored video frame based on a third search point displacement pattern is greater than the second search point displacement pattern; and d) from the first series of comparisons, the second series of comparisons and the third series of comparisons, determining a match video block, further comprising establishing the first region, the second region, and the third region of the previously stored video frame to have a rectangular geometry.

Claim 9

A method for compressing video, the method comprising the steps of: a) receiving a present video frame that includes a plurality of video blocks for each video block of the plurality of video blocks; b) performing a first series of comparisons between the each video block of the present video frame and a comparison video block of a previously stored video frame, wherein, for each comparison of the first series of comparison video block is incrementally positioned within a first region of the previously stored video frame based on a first search point displacement pattern; c) performing a second series of comparisons between the each video block and the compassion video block, wherein, for each comparison of the second series of comparisons, the comparison video block is incrementally positioned within a second region of the previously stored video frame based on a second search point displacement pattern, and wherein the second search point displacement pattern is greater that the first search point displacement pattern; d) performing a third series of comparisons between the each video block and the comparison video block, wherein, for each comparison of the third series of comparisons, the comparison video block is incrementally positioned within a third region of the previously stored video frame based on a third search point displacement pattern, and wherein the third search point displacement pattern is greater than the second search point displacement pattern; e) from the first series of comparisons, the second series of comparisons, and the third series of comparisons, determining a best match video block; f) subtracting the best match video block from the each video block to produce a residual video block; g) evaluating the best match video block based on the each video block; and h) encoding the residual video block when the evaluation is favorable, further comprising encoding the each video block when the evaluation is not favorable, and wherein step (e) comprises the steps of: e1) determining a first best match video block from the first series of comparisons, the second series of comparisons and the third series of comparisons; e2) performing a fourth series of comparisons between the each video block and the comparison video block, wherein for each comparison of the fourth series of comparisons, the comparison video block is incrementally positioned within a region surrounding the first best match video block based on a fourth search point displacement pattern, and wherein the fourth search point displacement pattern is equal to or smaller than the search point displacement of the first search point displacement pattern; and e3) determining the best match video block from the fourth series of comparisons.

Claim 10

A method for compressing video, the method comprising the steps of: a) receiving a present video frame that includes a plurality of video blocks for each video block of the plurality of video blocks; b) performing a first series of comparisons between the each video block of the present video frame and a comparison video block of a previously stored video frame, wherein, for each comparison of the first series of comparison video block is incrementally positioned within a first region of the previously stored video frame based on a first search point displacement pattern; c) performing a second series of comparisons between the each video block and the compassion video block, wherein, for each comparison of the second series of comparisons, the comparison video block is incrementally positioned within a second region of the previously stored video frame based on a second search point displacement pattern, and wherein the second search point displacement pattern is greater that the first search point displacement pattern; d) performing a third series of comparisons between the each video block and the comparison video block, wherein, for each comparison of the third series of comparisons, the comparison video block is incrementally positioned within a third region of the previously stored video frame based on a third search point displacement pattern, and wherein the third search point displacement pattern is greater than the second search point displacement pattern; e) from the first series of comparisons, the second series of comparisons, and the third series of comparisons, determining a best match video block; f) subtracting the best match video block from the each video block to produce a residual video block; g) evaluating the best match video block based on the each video block; and h) encoding the residual video block when the evaluation is favorable, further comprising encoding the each video block when the evaluation is not favorable, and wherein step (e) comprises the steps of: e1) determining a first best match video block from the first series of comparisons, the second series of comparisons and the third series of comparisons; e2) performing a fourth series of comparisons between the each video block and the comparison video block, wherein for each comparison of the fourth series of comparisons, the comparison video block is incrementally positioned within a region surrounding the first best match video block based on a fourth search point displacement pattern, and wherein the fourth search point displacement pattern is equal to or smaller than the search point displacement of the first search point displacement pattern; and e3) determining the best match video block from the fourth series of comparisons.