Coding of digital video with high motion content

• US 20020159529A1
• Filed: 11/15/2001
• Published: 10/31/2002
• Est. Priority Date: 03/15/2000
• Status: Active Grant

• Current Assignee: Etiip Holdings Incorporated
• Original Assignee: Dac International Inc.
• Inventors: Wang, Jian, Wang, Meng, Qu, Li
• Granted Patent: US 7,068,720 B2
• Time in Patent Office: Days
• US Class Current: 375/240.22
• CPC Class Codes: H04N 19/51 Motion estimation or motion compensation
• CPC Class Codes: H04N 19/94 Vector quantisation

Abstract

Methods for coding the current frame of a video data bit stream are provided. Methods include, for example, creating and maintaining a global codebook of historic values for the macroblocks of previous frames; searching entries in the global codebook to find the best estimation of the current frame and calculate a measure of the difference between values for the two macroblocks; comparing the effectiveness of the estimation macroblock provided by the global codebook entry to the estimation provided by a motion compensation scheme; choosing the best estimation method and coding the current frame; and updating the global codebook based on the new coded macroblock.

Claims

1. A method for coding the current frame of a video data bit stream comprising the steps of:

a) creating and maintaining a global codebook of historic values for the macroblocks of previous frames;

b) searching entries in the global codebook to find the best estimation of the current frame and calculate a measure of the difference between the values for the two macroblocks;

c) comparing the effectiveness of the estimation macroblock provided by the global code book entry to the estimation provided by a motion compensation scheme;

d) choosing the best estimation method and coding the current frame;

e) updating the global codebook based on the new coded macroblock.

2. The method of claim 1 wherein the entries in the global codebook are comprised of a data structure composed of data sets for luminance, chrominance, frame number and population.

3. The method of claim 1 wherein the global codebook is initially composed during a training mode using non-random data derived from sample video content.

4. The method of claim 1 wherein the macroblock is coded using neither the global codebook entry nor traditional motion compensation but rather using the mean values of all luminance and chrominance values for the current macroblock.

5. The method of claim 1 wherein the global codebook is updated by means of attributing increased importance to recently reconstructed macroblocks as opposed to older historic macroblocks.

6. The method of claim 1 wherein the global codebook is updated using a merged value of the current macroblock and the value of the original codebook entry.

7. The method of claim 1 wherein the global codebook is updated using the new absolute values for the macroblock rather than macroblock values derived from the merger of the new value and the original codebook entry.

8. An apparatus for coding video data using a compression encoder comprising the following components:

a) a global codebook of historic values for the macroblocks of previous frames;

b) means to search all entries in the global codebook to find the best estimation of the current frame and calculate a value for the difference between the two macroblocks;

c) means to compare the effectiveness of the estimation macroblock provided by the global code book entry to the estimation provided by the standard, prior art MPEG or H.263 motion compensation schemes;

d) means to determine the best estimation method

e) means to encode the current macroblock using the optimal predictive macroblock;

f) means of updating the global codebook using the newly encoded macroblocks.

9. The apparatus of claim 8 wherein the entries in the global codebook are comprised of a data structure composed of data sets for luminance, chrominance, frame number and population.

10. The apparatus of claim 8 wherein the global codebook is initially composed during a training mode using non-random data derived from sample video content.

11. The apparatus of claim 8 wherein macroblocks are coded using neither the global codebook entry nor traditional motion compensation but rather using the mean values of all luminance and chrominance values for the current macroblock.

12. The apparatus of claim 8 wherein the global codebook is updated by means of attributing increased importance to recently reconstructed macroblocks as opposed to older historic macroblocks.

13. The apparatus of claim 8 wherein the global codebook is updated using a merged value of the current macroblock and the value of the original codebook entry.

14. The apparatus of claim 8 wherein the global codebook is updated using the new absolute values for the macroblock rather than macroblock values derived from the merger of the new value and the original codebook entry.