Decoding method and apparatuses with candidate motion vectors
Various embodiments for deriving a motion vector predictor for a current block are provided. One or more first candidates for the motion vector predictor for the current block is derived from first motion vectors of first blocks. It is determined whether a total number of the one or more first candidates is less than a maximum candidate number. A second candidate for the motion vector predictor for the current block is derived when the total number of the one or more first candidates is less than the maximum candidate number, with the second candidate including a zero vector. The motion vector predictor for the current block is selected from among a plurality of candidates including the one or more first candidates and the second candidate.
1. A method for decoding a current block, the method comprising:
deriving one or more candidates of a first type for a motion vector predictor for the current block from first motion vectors of first blocks;
deriving a candidate of a second type for the motion vector predictor for the current block, the candidate of the second type being different from the one or more candidates of the first type and including a zero vector;
selecting the motion vector predictor for the current block from among a plurality of candidates including the one or more candidates of the first type and the candidate of the second type;
predicting the current block using the selected motion vector predictor for the current block; and
generating an image using the predicted current block,
wherein a total number of the one or more candidates of the first type is less than a maximum candidate number.
2. A decoding apparatus which decodes a current block, the decoding apparatus comprising:
a processor; and
a non-transitory storage,
the processor performing, using the non-transitory storage, processes including:
deriving one or more candidates of a first type for a motion vector predictor for the current block from first motion vectors of first blocks stored in the non-transitory storage;
deriving a candidate of a second type for the motion vector predictor for the current block, the candidate of the second type being different from the one or more candidates of the first type and including a zero vector;
selecting the motion vector predictor for the current block from among a plurality of candidates including the one or more candidates of the first type and the candidate of the second type;
predicting the current block using the selected motion vector predictor for the current block; and
generating an image using the predicted current block,
wherein a total number of the one or more candidates of the first type is less than a maximum candidate number.