MC16x16B

Performs motion compensation for bi-predicted block.

Syntax

IppStatus ippiMC16x16B_8u_C1(const Ipp8u* pSrcRefF, Ipp32s srcStepF, Ipp32s mcTypeF, const Ipp8u* pSrcRefB, Ipp32s srcStepB, Ipp32s mcTypeB, const Ipp16s* pSrcYData, Ipp32s srcYDataStep, Ipp8u* pDst, Ipp32s dstStep, Ipp32s roundControl);

Parameters

pSrcRefF	Pointer to the forward reference block.
srcStepF	Size of the row in bytes, specifying the aligned forward reference frame width.
mcTypeF	Forward MC type IPPVC_MC_APX.
pSrcRefB	Pointer to the backward reference block.
srcStepB	Size of the ow in bytes, specifying the aligned backward reference frame width.
mcTypeB	Backward MC type IPPVC_MC_APX.
pSrcYData	Pointer to the data obtained after inverse DCT.
srcYDataStep	Number of bytes, specifying the width of the aligned data obtained after inverse DCT.
pDst	Pointer to the destination predicted block.
dstStep	Size of the row in bytes, specifying the aligned destination frame width.
roundControl	Parameter that determines type of rounding for half a pel approximation; may be 0 or 1.

Description

This function is declared in the ippvc.h header file. The function ippiMC16x16B_8u_C1 calculates sum of 16x16 residual block and 16x16 predicted block, which is calculated as average of two 16x16 predictions (F-prediction and B-prediction) (See Figure Motion Compensation Scheme for Bi-Predicted Block). Each prediction is calculated on the basis of the corresponding reference block and corresponding mcType (see IPPVC_MC_APX).

Return Values

ippStsNoErr	Indicates no error.
ippStsNullPtrErr	Indicates an error when at least one input pointer is NULL.