Evaluates sum of squares of differences between current bi-predicted 16X16 block and mean of two reference blocks.
IppStatus ippiSqrDiff16x16B_8u32s(const Ipp8u* pSrc, Ipp32s srcStep, const Ipp8u pRefF, Ipp32s refStepF, Ipp32s mcTypeF, const Ipp8u pRefB, Ipp32s refStepB, Ipp32s mcTypeB, Ipp32s* pSqrDiff);
pSrc |
Pointer to the current block of specified size. |
srcStep |
Step of the current block, specifying width of the block in bytes. |
pRefF |
Pointer to the forward reference block of specified size. |
refStepF |
Step of the forward reference block, specifying width of the block in bytes. |
mcTypeF |
MC type IPPVC_MC_APX for the forward reference block. |
pRefB |
Pointer to the backward reference block of specified size. |
refStepB |
Step of the backward reference block, specifying width of the block in bytes. |
mcTypeB |
MC type IPPVC_MC_APX for the backward reference block. |
pSqrDiff |
Pointer to the sum of square difference between all elements in the current and reference blocks. |
This function is declared in the ippvc.h header file. The function ippiSqrDiff16x16B_8u32s evaluates the sum of square difference between all the elements in the current block and the mean of corresponding elements in two reference blocks of 8x8 elements. The result is stored in integer pSqrDiff.
Let us denote the pel which lies at the crossing of ith row and jth column of the current block as block[i,j], the pel which lies at the crossing of ith row and jth column of the forward reference block as f_block[i,j], and the pel which lies at the crossing of ith row and jth column of the forward reference block as b_block[i,j]. Then
ippStsNoErr |
Indicates no error. |
ippStsNullPtrErr |
Indicates an error when at least one input pointer is NULL. |
ippStsStepErr |
Indicates an error when srcStep, refStepF or refStepB is less than or equal to zero. |
Copyright © 2000 - 2011, Intel Corporation. All rights reserved.