Method and apparatus for scalable motion estimation
View Patent ↗Embodiments of the invention generally provide a method and apparatus for scalable motion estimation. One embodiment of a method for performing motion estimation includes setting a target complexity for a macroblock of a source frame and performing motion estimation for one or more partitions of the macroblock until the target complexity is met.
1. A method for performing motion estimation, comprising:
at a motion estimator:
setting a target complexity for a macroblock of a source frame, wherein the target complexity represents a maximum number of motion estimation searches that are allowed for the macroblock; and
performing, by at least one processor or circuit, motion estimation for at least a subset of a plurality of partitions of the macroblock containing more than one type of partition, including at least one rectangular square-shaped partition and at least one partition of a different shape or an order of decreasing size, wherein after finishing motion estimation for a partition in the plurality of partitions of the macroblock, motion estimation for another partition in the plurality of partitions of the macroblock is performed until the target complexity is met.
2. The method of claim 1 , wherein the target complexity is derived from a metric based on a sum of absolute differences count for the macroblock.
3. The method of claim 1 , wherein the motion estimation is performed on at least one type of the plurality of partitions that is square-shaped, while at least one other type of the plurality of partitions is not square-shaped.
4. The method of claim 3 , wherein the motion estimation is performed for the plurality of partitions that are square-shaped, by types of the plurality of partitions in order of decreasing size.
5. The method of claim 3 , wherein the motion estimation is performed using at least one reference picture for each of the at least one of the plurality of partitions that is square-shaped.
6. The method of claim 5 , wherein the at least one reference picture comprises one or more reference pictures for a predictive pictures or two or more reference pictures for a bi-predictive picture.
7. The method of claim 1 , further comprising:
setting a motion estimation cost for all of the one or more partitions to a maximum possible value.
8. The method of claim 7 , wherein the maximum possible value is a highest possible value allowed by a particular hardware or software implementation of the method.
9. The method of claim 1 , wherein:
when partitions of the same type are used, a different number of reference pictures is used to perform the motion estimation.
10. A method for performing motion estimation, comprising:
at a motion estimator: setting a target complexity for a macroblock of a source frame, wherein the target complexity represents a maximum number of motion estimation searches that are allowed for the macroblock; and
performing, by at least one processor or circuit, motion estimation for at least a subset of a plurality of partitions of the macroblock,
wherein after finishing motion estimation for a partition in the plurality of partitions of the macroblock, motion estimation for another partition in the plurality of partitions of the macroblock is performed until the target complexity is met, and
wherein the motion estimation is performed on at least one of the plurality of partitions that is rectangular-shaped, only if the target complexity is not met after performing the motion estimation for at least one of the plurality of partitions that is square-shaped.
11. The method of claim 10 , wherein the motion estimation is performed using at least one reference picture for each of the at least one of the plurality of partitions that is rectangular-shaped.
12. The method of claim 1 , wherein the motion estimation comprises a simplified hexagon search.
13. The method of claim 11 , wherein the at least one reference picture comprises one or more reference pictures for a predictive pictures or two or more reference pictures for a bi-predictive picture.
14. A non-transitory computer readable storage medium containing an executable program for motion performing estimation, where the program performs the steps of:
setting a target complexity for a macroblock of a source frame, wherein the target complexity represents a maximum number of motion estimation searches that are allowed for the macroblock; and
performing, by at least one processor or circuit, motion estimation for at least a subset of a plurality of partitions of the macroblock containing more than one type partition, including at least one rectangular square-shaped partition and at least one partition of a different shape or an order of decreasing size, wherein after finishing motion estimation for a partition in the plurality of partitions of the macroblock, motion estimation for another partition in the plurality of partitions of the macroblock is performed until the target complexity is met.
15. The non-transitory computer readable storage medium of claim 14 , wherein the motion estimation is performed on at least one type of the plurality of partitions that is square-shaped, while at least one other type of the plurality of partitions is not square-shaped.
16. The non-transitory computer readable storage medium of claim 15 , wherein the motion estimation is performed using at least one reference picture for each of the at least one of the plurality of partitions that is square-shaped.
17. The non-transitory computer readable storage medium of claim 15 , wherein the motion estimation is performed on at least one of the plurality of partitions that is rectangular-shaped, only if the target complexity is not met after performing the motion estimation for the at least one of the plurality of partitions that is square-shaped.
18. The non-transitory computer readable storage medium of claim 17 , wherein the motion estimation is performed using at least one reference picture for each of the at least one of the plurality of partitions that is square-shaped.
19. A method for performing motion estimation, comprising:
at a motion estimator:
setting a target complexity for a macroblock of a source frame, wherein the target complexity represents a maximum number of motion estimation searches that are allowed for the macroblock; and
performing, by at least one processor or circuit, motion estimation for at least a subset of a plurality of partitions of the macroblock, where the performing motion estimation comprises:
performing motion estimation for a first square-shaped partition of the macroblock and for a first number of reference pictures;
performing motion estimation for a second square-shaped partition of the macroblock that is smaller in size than the first square-shaped partition and for the first number of reference pictures;
performing motion estimation for the first square-shaped partition and for a second number of reference pictures;
performing motion estimation for the second square-shaped partition and for the second number of reference pictures;
performing motion estimation for a first rectangular-shaped partition of the macroblock and for a the first number of reference pictures;
performing motion estimation for a second rectangular-shaped partition of the macroblock and for a the first number of reference pictures;
performing motion estimation for the first rectangular-shaped partition and for the first number of reference pictures; and
performing motion estimation for the second rectangular-shaped partition and for the first number of reference pictures,
wherein after each motion estimation for one of the partitions is performed, determining if the target complexity is met and terminating the performing of motion estimation when the target complexity is met.