Potential issue in determining vertical luma interpolation filter coefficients

[vdrugeon]

In 8.5.6.3.2 Luma sample interpolation filtering process, when calculating vertical luma interpolation filter coefficients:
"The vertical luma interpolation filter coefficients fLV[ p ] for each 1/16 fractional sample position p equal to yFracL are derived as follows:

– If MotionModelIdc[ xSb ][ ySb ] is greater than 0, and sbWidth and sbHeight are both equal to 4, and scalingRatio[ 0 ] is greater than 28672, the luma interpolation filter coefficients fLV[ p ] are specified in Table 31.
– Otherwise, if MotionModelIdc[ xSb ][ ySb ] is greater than 0, and sbWidth and sbHeight are both equal to 4, and scalingRatio[ 0 ] is greater than 20480, the luma interpolation filter coefficients fLV[ p ] are specified in Table 32.
– Otherwise, if MotionModelIdc[ xSb ][ ySb ] is greater than 0, and sbWidth and sbHeight are both equal to 4, the luma interpolation filter coefficients fLV[ p ] are specified in Table 30.
– Otherwise, if scalingRatio[ 1 ] is greater than 28672, luma interpolation filter coefficients fLV[ p ] are specified in Table 28.
– Otherwise, if scalingRatio[ 1 ] is greater than 20480, luma interpolation filter coefficients fLV[ p ] are specified in Table 29.
– Otherwise, the luma interpolation filter coefficients fLV[ p ] are specified in Table 27 depending on hpelIfIdx set equal to hpelVerIfIdx."

Is it intended that scalingRatio[0], i.e. the horizontal scaling factor, is used for determining the vertical filter coefficients in the first two bullets, but scalingRatio[1], i.e. the vertical scaling factor, is used in subsequent bullets? Could someone familiar with this process please check?

[jlchen]

They should be scalingRatio[1]. will be fixed in vE release.