Method of decoding moving picture
View Patent ↗Provided is a method that generates a prediction block of a current prediction unit, and generates a residual block of the current prediction unit. To generate the prediction block, a reference picture index and motion vector difference of the current prediction unit are obtained from a received bit stream, spatial and temporal motion vector candidates are derived to construct a motion vector candidate list. A motion vector predictor is selected and a motion vector of the current prediction unit is restored to generate the prediction block. Therefore, the motion vector encoded effectively using spatial and temporal candidates is correctly recovered and the complexity of a decoder is reduced.
1. A method of decoding a moving picture, comprising:
a step of generating a prediction block of a current prediction unit; and
a step of generating a residual block of the current prediction unit,
wherein the step of generating the prediction block of the current prediction unit includes the steps of:
obtaining a reference picture index and motion vector difference of the current prediction unit from a received bit stream;
constructing a motion vector candidate list using available spatial and temporal motion vector candidates;
determining a motion vector candidate indicated by a motion vector index as a motion vector predictor and restoring a motion vector of the current prediction unit using the motion vector difference and the motion vector predictor; and
generating a prediction block of the current prediction unit using the restored motion vector and the reference picture index,
wherein the temporal motion vector candidate is a first available motion vector encountered when retrieving two blocks corresponding to the current prediction unit or a motion vector of a predetermined block,
wherein the step of generating the residual block of the current prediction unit includes the steps of:
entropy-decoding residual signals;
inversely scanning the residual signals using a first scan pattern;
restoring a transform block by inversely quantizing a quantized transform block; and
inversely transforming the transform block, and
wherein, when a size of the transform block is larger than a first reference size, the residual signals are inversely scanned in a unit of subsets and a scan pattern for scanning a plurality of subsets is the same as a scan pattern for scanning residual signals of each subset.
2. The method of claim 1 , wherein the step of generating the prediction block of the current prediction unit further includes the steps of:
restoring a reference picture index and a differential motion vector of the current prediction unit from a received bit stream;
constructing a motion vector candidate list using available spatial and temporal motion vector candidates;
determining a motion vector candidate indicated by a motion vector index as a motion vector predictor; and
restoring a motion vector of the current prediction unit using the differential motion vector and the motion vector predictor.
3. The method of claim 2 , wherein, when the current prediction unit is located at a below border of a current maximum coding unit, the temporal motion vector candidate is a motion vector of a center block including an above-left pixel or a below-left pixel of center position of a block corresponding to the current prediction unit within a temporal motion vector candidate picture.
4. The method of claim 2 , wherein, when a left spatial motion vector of the current prediction unit is scaled, an above spatial motion vector of the current prediction unit to be scaled is set to unavailable.