FIELD: information technology.
SUBSTANCE: invention relates to encoding and decoding video data. Disclosed is a method of decoding an image. Method includes a step of obtaining prediction mode information of current unit from a bit stream, and obtaining motion vector predictor candidates of current unit using a unit, jointly arranged with current unit in a reference frame in a temporal direction, based on obtained prediction mode information. Further, method comprises obtaining a motion vector predictor of current unit from among obtained motion vector predictor candidates based on information, indicating a motion vector predictor, used for current unit.
EFFECT: high efficiency of decoding images by setting size of coding unit.
1 cl, 27 dwg
