Fwd: wrong colors with Xv extension and image format id: 0x59565955 (UYVY) from an rgb mapping

[Tomas Carnecky]

On 01/23/2009 02:59 AM, Amos Tibaldi wrote:

> #define GUID_UYVY_PLANAR 0x59565955

Just a sidenote: This is misleading, that format is packed and not planar.

> void RGBToUV(unsigned short int r,
> unsigned short int g,
> unsigned short int b,
> unsigned short int * u,
> unsigned short int * v)
> {
> *u = -0.147 * r +
> -0.289 * g +
> 0.436 * b; // min(abs(r * -1214 + g * -2384 + b * 3598 + 4096 + 1048576)
>  >> 13, 240);
> //(unsigned short int)(-0.147f*(float)r-0.289f*(float)g+0.436f*(float)b);
> *v = 0.615 * r +
> -0.515 * g +
> -0.100 * b;
> //min(abs(r * 3598 + g * -3013 + b * -585 + 4096 + 1048576) >> 13, 240);
> //(unsigned short int)(0.615f*(float)r-0.515f*(float)g-0.100f*(float)b);
> }
> void RGBToY(unsigned short int r,
> unsigned short int g,
> unsigned short int b,
> unsigned short int * y
> )
> {
> *y = 0.299 * r +
> 0.587 * g +
> 0.114 * b; // min(abs(r * 2104 + g * 4130 + b * 802 + 4096 + 131072) >>
> 13, 235);
> //(unsigned short int)(0.299f*(float)r+0.587f*(float)g+0.114f*(float)b);
> }

The formulas that are in use are wrong. You need to scale and shift the 
YUV values. The formulas that you commented out are correct and yield 
the correct colors here (min(abs(...) >> 13, 235)). These formulas also 
worked in my test:
*y = ( (  66 * r + 129 * g +  25 * b + 128) >> 8) +  16;
*u = ( ( -38 * r -  74 * g + 112 * b + 128) >> 8) + 128;
*v = ( ( 112 * r -  94 * g -  18 * b + 128) >> 8) + 128;

See wikipedia or msdn:
http://msdn.microsoft.com/en-us/library/ms893078.aspx
http://en.wikipedia.org/wiki/YUV#Numerical_approximations

tom