Scalable implementation for parallel motion estimation regions

A method of encoding video data, the method comprising:

• determining information associated with a plurality of motion vector candidates, each of the plurality of motion vector candidates corresponding to at least one of a plurality of prediction units (PUs) partitioned in a parallel motion estimation region (MER), each of the plurality of motion vector candidates comprising available motion data;
• selecting a subset of the plurality of motion vector candidates to include in a merge candidate list, wherein selection is based on a priority level of each of the plurality of motion vector candidates, wherein the priority level of each of the plurality of motion vector candidates is based on (i) a frequency for which the motion vector candidate corresponds to any of the plurality of PUs partitioned in the MER such that a motion vector candidate that corresponds to a higher number of PUs is associated with a higher priority level and (ii) a weighting factor based on a size of each PU such that a motion vector candidate that corresponds to a larger PU is associated with a higher priority level;
• and
• generating the merge candidate list to include the selected subset of the plurality of motion vector candidates.

A video coding apparatus, the apparatus comprising:

• means for determining information associated with a plurality of motion vector candidates, each of the plurality of motion vector candidates corresponding to at least one of a plurality of prediction units (PUs) partitioned in a parallel motion estimation region (MER), each of the plurality of motion vector candidates comprising available motion data;
• means for selecting a subset of the plurality of motion vector candidates to include in a merge candidate list, wherein selection is based on a priority level of each of the plurality of motion vector candidates, wherein the priority level of each of the plurality of motion vector candidates is based on (i) a frequency for which the motion vector candidate corresponds to any of the plurality of PUs partitioned in the MER such that a motion vector candidate that corresponds to a higher number of PUs is associated with a higher priority level and (ii) a weighting factor based on a size of each PU such that a motion vector candidate that corresponds to a larger PU is associated with a higher priority level;
• and
• means for generating the merge candidate list to include the selected subset of the plurality of motion vector candidates.

A non-transitory computer readable medium having stored thereon code that when executed causes an apparatus to:

• receive information associated with a plurality of motion vector candidates, each of the plurality of motion vector candidates corresponding to at least one of a plurality of prediction units (PUs) partitioned in a parallel motion estimation region (MER), each of the plurality of motion vector candidates comprising available motion data;
• select a subset of the plurality of motion vector candidates to include in a merge candidate list, wherein selection is based on a priority level of each of the plurality of motion vector candidates, wherein the priority level of each of the plurality of motion vector candidates is based on (i) a frequency for which the motion vector candidate corresponds to any of the plurality of PUs partitioned in the MER such that a motion vector candidate that corresponds to a higher number of PUs is associated with a higher priority level and (ii) a weighting factor based on a size of each PU such that a motion vector candidate that corresponds to a larger PU is associated with a higher priority level;
• and
• generate the merge candidate list to include the selected subset of the plurality of motion vector candidates.