Motion estimation in enhancement layers in video encoding

1. A method for encoding a video sequence in a scalable video encoder, the method comprising:

selecting a first search window configuration for coding a current picture of the video sequence in an enhancement layer encoder of the scalable video encoder based on motion vectors generated by a reference layer encoder when encoding the current picture, wherein selecting a first search window configuration comprises determining motion vector coverage for the first search window configuration based on the motion vectors, comprising:

scaling a motion vector; and

adding a cost of the motion vector to a cumulative motion vector cost if the scaled motion vector fits in the first search window configuration and the cost is less than a cost threshold,

wherein the cumulative motion vector cost indicates the motion vector coverage of the first search window configuration; and

coding the current picture in the enhancement layer encoder using the first search window configuration.

2. The method of claim 1 , wherein the reference layer encoder is a base layer encoder in the scalable video encoder.

3. The method of claim 1 , wherein the first search window configuration is selected from a plurality of search window configurations comprising a growing search window configuration and a hybrid search window configuration.

4. The method of claim 1 , wherein the current picture is encoded in the reference layer encoder at a lower resolution than in the enhancement layer encoder.

5. The method of claim 1 , further comprising

determining a global motion vector for the current picture when the current picture is encoded by the reference layer encoder; and

wherein coding the current picture in the enhancement layer encoder further comprises coding a macroblock in the current picture, wherein at least one of the global motion vector and a motion vector of a collocated macroblock are used as initial search points for motion estimation.

6. The method of claim 1 , further comprising selecting a second search window configuration for coding another picture of the video sequence in the enhancement layer encoder based on motion vectors generated by the reference layer encoder when encoding the another picture, wherein the second search window configuration is different from the first search window configuration.

7. The method of claim 1 , wherein selecting a first search window configuration further comprises:

selecting the first search window configuration from a plurality of search window configurations when the motion vector coverage of the first search window configuration is better than motion vector coverage of other search window configurations in the plurality of search window configurations.

8. The method of claim 7 , wherein determining motion vector coverage comprises:

scaling a motion vector; and

incrementing a motion vector count if the scaled motion vector fits in the first search window configuration,

wherein the motion vector count indicates the motion vector coverage of the first search window configuration.

9. The method of claim 7 , wherein the motion vectors are one selected from a group consisting of all motion vectors generated by the reference layer encoder when encoding the current picture and motion vectors of macroblocks in the current picture that were inter coded by the reference layer encoder.

10. A digital system comprising a video encoder for encoding a video sequence, the video encoder comprising:

means for selecting a first search window configuration for coding a current picture of the video sequence in an enhancement layer encoder of the scalable video encoder based on motion vectors generated by a reference layer encoder when encoding the current picture, wherein the means for selecting a first search window configuration comprises means for determining motion vector coverage for the first search window configuration based on the motion vectors, comprising:

means for scaling a motion vector; and

means for adding a cost of the motion vector to a cumulative motion vector cost if the scaled motion vector fits in the first search window configuration and the cost is less than a cost threshold,

wherein the cumulative motion vector cost indicates the motion vector coverage of the first search window configuration; and

means for coding the current picture in the enhancement layer encoder using the first search window configuration.

11. The digital system of claim 10 , wherein the reference layer encoder is a base layer encoder in the scalable video encoder.

12. The digital system of claim 10 , wherein the first search window configuration is selected from a plurality of search window configurations comprising a growing search window configuration and a hybrid search window configuration.

13. The digital system of claim 10 , wherein the current picture is encoded in the reference layer encoder at a lower resolution than in the enhancement layer encoder.

14. The digital system of claim 10 , further comprising means for selecting a second search window configuration for coding another picture of the video sequence in the enhancement layer encoder based on motion vectors generated by the reference layer encoder when encoding the another picture, wherein the second search window configuration is different from the first search window configuration.

15. The digital system of claim 10 , wherein the means for selecting a first search window configuration further comprises:

means for selecting the first search window configuration from a plurality of search window configurations when the motion vector coverage of the first search window configuration is better than motion vector coverage of other search window configurations in the plurality of search window configurations.

16. The digital system of claim 15 , wherein the means for determining motion vector coverage comprises:

means for scaling a motion vector; and

means for incrementing a motion vector count if the scaled motion vector fits in the first search window configuration,

wherein the motion vector count indicates the motion vector coverage of the first search window configuration.

17. The digital system of claim 15 , wherein the motion vectors are one selected from a group consisting of all motion vectors generated by the reference layer encoder when encoding the current picture and motion vectors of macroblocks in the current picture that were inter coded by the reference layer encoder.