Method for motion vector coding of MPEG-4

• US 6,498,810 B1
• Filed: 04/24/1998
• Issued: 12/24/2002
• Est. Priority Date: 09/12/1997
• Status: Expired due to Term

• Current Assignee: LG Electronics, Inc. (LG Corporation)
• Original Assignee: LG Electronics, Inc. (LG Corporation)
• Inventors: Kim, Hyun Mun, Ra, Jong Beom, Kim, Sung Deuk, Lee, Young Su
• Primary Examiner(s): Kelley, Chris
• Assistant Examiner(s): An, Shawn S.
• Application Number:
  US09/065,578
• Time in Patent Office: 1,705 Days
• Field of Search: 375/240.16, 375/240, 375/240.24, 348/699, 348/416, 348/413, 348/243, 348/405, 382/232, 382/236, 382/238
• US Class Current: 375/240
• CPC Class Codes: H04N 19/51 Motion estimation or motion compensation
• CPC Class Codes: H04N 19/517 by encoding
• CPC Class Codes: H04N 19/567 Motion estimation based on rate distortion criteria

Abstract

A method for motion vector coding of an MPEG-4 includes determining motion vector candidates, and selecting a candidate having a minimum bit rate among motion vector candidates in an encoder. A motion vector prediction error and prediction mode information and then to a decoder. The decoder estimates a minimum bit rate for coding a motion vector difference and codes a motion vector against components X and Y, using a minimum rate predictor.

Claims

1. A method for motion vector coding comprising:

selecting a candidate having a minimum coding bit rate from a plurality of motion vector candidates for a macroblock;

sending a motion vector prediction error and prediction mode information for the macroblock to a decoder;

estimating a minimum bit rate for coding a motion vector difference of the macroblock in the decoder; and

coding a motion vector against a first component using a minimum rate predictor in the decoder, wherein the motion vector prediction error determines a subset of the plurality of motion vector candidates, and wherein the prediction mode information determines one of the subset of the plurality of candidates used in coding the motion vector difference for coding a motion vector.

2. The method according to claim 1, wherein coding the motion vector against the first component uses the following equation:

3. The method according to claim 1, further comprising coding a motion vector against a second component using the minimum rate predictor, wherein the first and second components are X and Y components.

4. The method according to claim 1, wherein the prediction mode information designates the selected candidate having the minimum coding bit rate.

5. The method according to claim 1, wherein the sending step further comprises:

subtracting a motion vector corresponding to each of the plurality of candidates from a current motion vector; and

selecting the smallest value as the motion vector prediction error.

6. The method according to claim 1, wherein the estimating a minimum bit rate for coding a motion vector difference step comprises selecting the motion vector of the candidate motion vectors that corresponds to the minimum rate predictor.

7. The method according to claim 1, wherein the motion vector prediction error and the prediction mode information indicate which single one of the plurality of motion vector candidates to use in decoding the macroblock.

8. The method according to claim 1, wherein the decoding a motion vector against a first component comprises:

determining a rate predictor value for each of the plurality of motion vector candidates;

determining a minimum rate predictor among the plurality of rate predictor values;

setting a minimum rate predictor value to the corresponding motion vector candidate according to the minimum rate predictor for the first component;

setting a motion vector difference minimum rate using the relationship between the motion vector of the current block and-the minimum rate predictor value; and

coding the motion vector difference using the motion vector difference minimum rate.

9. The method according to claim 1, wherein the estimating and the coding in the decoder use the motion vector prediction error and the prediction mode information determined in an encoder.

10. A method for motion vector coding comprising:

selecting a candidate having a minimum coding bit rate from a plurality of motion vector candidates for a macroblock;

sending a motion vector prediction error and prediction mode information for the macroblock to a decoder;

estimating a minimum bit rate for coding a motion vector difference of the macroblock in the decoder; and

coding a motion vector against a first component using a minimum rate predictor in the decoder, wherein coding the motion vector against the first component uses the following equation;

11. A method for motion vector encoding comprising:

identifying a plurality of motion vector candidates for a current block of a macroblock;

selecting one of a minimum rate prediction mode and a median prediction mode for the current block;

selecting a candidate having a minimum coding bit rate from the plurality of motion vector candidates for the minimum rate prediction mode;

determining a motion vector prediction error and prediction mode information as encoded information for the minimum rate prediction mode;

determining the encoded information for the median prediction mode; and

transmitting the encoded information, wherein the minimum rate prediction mode selects one of the plurality of motion vector candidates to decode the current block, and wherein the median prediction mode uses a combination of the plurality of motion vector candidates as a motion vector to decode the current block.

12. The method according to claim 11, wherein the determining a motion vector prediction error step comprises:

subtracting a motion vector corresponding to each of the plurality of candidates from the current motion vector; and

selecting the smallest value as the motion vector prediction error.

13. The method according to claim 11, wherein the prediction mode information identifies a selected candidate of the plurality of candidates.

14. The method according to claim 11, wherein the prediction mode information selects one of the plurality of motion vector candidates used in coding a motion vector difference.

15. The method according to claim 11, wherein the transmitted motion vector prediction error and prediction mode information is used in coding the motion vector against the first component using a minimum rate predictor determined by the following equation:

16. A method for motion vector decoding comprising:

receiving a motion vector prediction error and prediction mode information for a macroblock as encoded information for a minimum rate prediction mode in a decoder;

estimating a minimum bit rate for coding a motion vector difference for the macroblock in the decoder; and

coding a motion vector for the current block against a first component using a minimum rate predictor for the macroblock in the decoder, wherein the motion vector prediction error determines a subset of the plurality of motion vector candidates, and wherein the prediction mode information determines one of the subset of the plurality of candidates used in coding the motion vector difference.

17. The method according to claim 16, further comprising coding a motion vector against a second component using the minimum rate predictor, wherein the first and second components are X and Y components.

18. The method according to claim 16, wherein the prediction mode information designates a selected candidate having the minimum coding bit rate.

19. The method according to claim 16, wherein the estimating a minimum bit rate for coding a motion vector difference step comprises selecting the motion vector of the candidate motion vectors that corresponds to the minimum rate predictor.

20. The method according to claim 16, further comprising coding the motion vector for the current block against the first component using a median prediction motion vector for a median prediction mode.

21. The method according to claim 16, wherein the decoding a motion vector against a first component comprises:

determining a rate predictor value for each of the plurality of motion vector candidates;

determining a minimum rate predictor among the plurality of rate predictor values;

setting the minimum rate predictor value to the corresponding motion vector candidate according to the minimum rate predictors for the first component;

setting a motion vector difference minimum rate using the relationship between the motion vector of the current block and the minimum rate predictor value;

and

coding the motion vector difference using the motion vector difference minimum rate.

22. The method according to claim 16, wherein the estimating and the coding in the decoder use the motion vector prediction error and the prediction mode information determined in an encoder.

23. A method for motion vector coding comprising:

selecting a candidate having a minimum coding bit rate from a plurality of motion vector candidates for a macroblock;

sending a motion vector prediction error and prediction mode information for the macroblock to a decoder;

estimating a minimum bit rate for coding a motion vector difference of the macroblock in the decoder; and

coding a motion vector against a first component using a minimum rate predictor in the decoder, wherein the coding a motion vector against a first component comprises,

determining a rate predictor value for each of the plurality of motion vector candidates,

determining a minimum rate predictor among the plurality of rate predictor values,

setting the minimum rate predictor value to the corresponding motion vector candidate according to the minimum rate predictor for the first component,

setting a motion vector difference minimum rate using the relationship between the motion vector of the current block and the minimum rate predictor value, and

coding the motion vector difference using the motion vector difference minimum rate.

24. A method for motion vector decoding comprising:

receiving a motion vector prediction error and prediction mode information for a macroblock as encoded information for a minimum rate prediction mode in a decoder;

estimating a minimum bit rate for coding a motion vector difference for the macroblock in the decoder; and

coding a motion vector for the current block against a first component using a minimum rate predictor for the macroblock in the decoder, wherein the decoding a motion vector against a first component comprises,

determining a rate predictor value for each of the plurality of motion vector candidates,

determining the minimum rate predictor among the plurality of rate predictor values,

setting a minimum rate predictor value to the corresponding motion vector candidate according to the minimum rate predictors for the first component,

setting a motion vector difference minimum rate using the relationship between the motion vector of the current block and the minimum rate predictor value, and

coding the motion vector difference using the motion vector difference minimum rate.