Averages two blocks with weights.
Syntax
IppStatus ippiWeightedAverage_H264_8u_C1IR(const Ipp8u* pSrc1, Ipp8u* pSrc2Dst, Ipp32s srcDstStep, Ipp32s weight1, Ipp32s weight2, Ipp32s shift, Ipp32s offset, IppiSize roiSize);
Parameters
pSrc1 |
Pointer to the first source (output of preceding functions). |
pSrc2Dst |
Pointer to the second source and result. |
srcDstStep |
Distance in bytes between starts of the consecutive lines in the source and result images. |
weight1, weight2 |
Weights. |
shift |
Shift. |
offset |
Offset. |
roiSize |
Flag that specifies the region of interest (could be 16, 8, 4 or 2 in each dimension). |
Description
The function ippiWeightedAverage_H264_8u_C1IR is declared in the ippvc.h file. This function averages two blocks with weights (for weighted bi-directional predictions) as specified in 8.4.2.3.2 of [JVTG050] when predFlagL0 and predFlagL1 are equal to 1.
The function uses the following formula:
where x ∈ [0,roi.width-1], y ∈ [0,roi.height-1]
The above formula corresponds to formula 8-220 of JVT-G050. See Table “Color Conversion Functions” for matching the function arguments and the standard variables.
| ippiWeightedAverage | 8.4.2.3.2 JVT-G050 |
|---|---|
| weight1 | w0 |
| weight2 | w1 |
| shift | logWD +1 |
| offset | (o0 + o1 + 1)>>1 |
| roi.width | partWidth |
| roi.height | partHeight |
Values of w0, w1, logWD, o0, o1 are calculated from formulas 8-221 through 8-244.
This function is used in the H.264 decoder included into Intel IPP Samples. See introduction to H.264.
Return Values
ippStsNoErr |
Indicates no error. |
ippStsNullPtrErr |
Indicates an error condition if at least one of the specified pointers is NULL. |
ippStsStepErr |
Indicates an error condition if srcDstStep value is less than roi.width |