Method and apparatus for derivation of motion vector predictor candidate set

An apparatus for deriving a motion vector predictor (MVP) candidate set for a block, the apparatus comprising one or more circuits configured to:

• determine a plurality of spatial neighboring blocks of the block;
• obtain one or more spatial MVP candidates from motion vectors associated with the spatial neighboring blocks;
• determine whether one or more redundant MVP candidates exist in the spatial MVP candidates;
• generate a first MVP candidate set, wherein said generating the first MVP candidate set comprises not including the determined one or more redundant MVP candidates into the first MVP candidate set;
• generate a final MVP candidate set according to the first MVP candidate set, wherein said generating the final MVP candidate set comprises:
• if a size of the first MVP candidate set is smaller than a target size, adding one or more pre-defined replacement MVP candidates to the first MVP candidate set until the size of the first MVP candidate set equal to the target size, and regarding the first MVP candidate set as the final MVP candidate set;
• and
• encode or decode the block using the final MVP candidate set;
• wherein a value is assigned to each pre-defined replacement MVP candidate,
• wherein when the block is encoded or to be encoded by Merge mode, the value corresponds to one or a combination of prediction direction, reference index, and motion vector, and said one or more pre-defined replacement MVP candidates added to the first MVP candidate set comprise both a first zero vector with a reference frame index equal to 0 in list 0 and a second zero vector with a reference frame index equal to 0 in list 1,
• wherein when the block is encoded or to be encoded by Advanced Motion Vector Prediction (AMVP) mode, the value corresponds to motion vector (MV), and said one or more pre-defined replacement MVP candidates added to the first MVP candidate set comprise one or more zero vectors.

A non-transitory computer readable medium storing a computer-executable program, the computer-executable program, when executed, causing a video coder to perform the following steps:

• determining a plurality of spatial neighboring blocks of the block;
• obtaining one or more spatial MVP candidates from motion vectors associated with the spatial neighboring blocks;
• determining whether one or more redundant MVP candidates exist in the spatial MVP candidates;
• generating a first MVP candidate set, wherein said generating the first MVP candidate set comprises not including the determined one or more redundant MVP candidates into the first MVP candidate set;
• generating a final MVP candidate set according to the first MVP candidate set, wherein said generating the final MVP candidate set comprises:
• if a size of the first MVP candidate set is smaller than a target size, adding one or more pre-defined replacement MVP candidates to the first MVP candidate set until the size of the first MVP candidate set equal to the target size, and regarding the first MVP candidate set as the final MVP candidate set;
• and
• encoding or decoding the block using the final MVP candidate set;
• wherein a value is assigned to each pre-defined replacement MVP candidate,
• wherein when the block is encoded or to be encoded by Merge mode, the value corresponds to one or a combination of prediction direction, reference index, and motion vector, and said one or more pre-defined replacement MVP candidates added to the first MVP candidate set comprise both a first zero vector with a reference frame index equal to 0 in list 0 and a second zero vector with a reference frame index equal to 0 in list 1,
• wherein when the block is encoded or to be encoded by Advanced Motion Vector Prediction (AMVP) mode, the value corresponds to motion vector (MV), and said one or more pre-defined replacement MVP candidates added to the first MVP candidate set comprise one or more zero vectors.