- 3 - INTER-VIEW RESIDUAL PREDICTION IN MULTI-VIEW OR 3-DIMENSIONAL VIDEO CODING

摘要

Video coder scaling the motion vector of the current prediction unit (PU) of the current picture in order to compensate for differences in the temporal distance. In addition, the video coder of determining the prediction block for the current PU, and determining the disparity of the reference blocks based on the display samples of the parity reference picture in the location indicated by the current PU disparity vector, and the scaled motion and by a vector based on samples of the fixed reference picture from the displayed location, in time for the current PU - determines the disparity of the reference blocks. Next, the video coder determines who prediction residuals for the current PU. Each sample's residual prediction for the current PU is time for the current PU - represents the difference between the samples corresponding to the display sample and the disparity of the reference blocks of the parity reference block.;