Image Processing - Practical Rendering and Computation with Direct3D 11 - page 472

the pixel location indicated by the dispatch thread ID in much the same way that was used
to read the appropriate data from the input image. This initial implementation is shown in
Listing 10.1 and visualized for a single pixel in Figure 10.7.
// Declare the input and output resources
Texture2D<float4> InputHap : register( t0 );
RWTexture2D<float4> OutputMap : register( u0 );
// Group size
#define size_x 32
#define size_y 32
// Declare the filter kernel coefficients
static const float filter[7][7] = {
0.000904706, 0.00315-7733, 0.00668492, 0.008583607, 0.00668492,
0.003157733, 0.000904706/
0.003157733, 0.01102157, 0.023332663, 0.029959733, 0.023332663,
0.01102157, 0.003157733,
0.00668492, 0.023332663, 0.049395249, 0.063424755, 0.049395249,
0.023332663, 0.00668492,
0.008583607, 0.029959733, 0.063424755, 0.081438997, 0.063424755,
0.029959733, 0.008583607,
0.00668492, 0.023332663, 0.049395249, 0.063424755, 0.049395249,
0.023332663, 0.00668492,
0.003157733, 0.01102157, 0.023332663, 0.029959733, 0.023332663,
0.01102157, 0.003157733,
0.000904706, 0.003157733, 0.00668492, 0.008583607, 0.00668492,
0.003157733, 0.000904706
};
II Declare one thread for each texel of the current block size.
[numthreads(size_x, size_y, 1)]
void CSMAIN( uint3 DispatchThreadID : SV_DispatchThreadID )
{
// Offset the texture location to the first sample location
int3 texturelocation = DispatchThreadID - int3( 3, 3, 0 );
// Initialize the output value to zero, then loop through the
// filter samples, apply them to the image samples, and sum
// the results.
float4 Color = float4( 0.0, 0.0, 0.0, 0.0 );
for ( int x = 0; x < 7; x++ )
for ( int y = 0; y < 7; y++ )
Color += InputMap.Load( texturelocation + int3(x,y,0) ) * filter[x][y];
// Write the output to the output resource
OutputMap[DispatchThreadID.xy] = Color;
}
Listing 10.1. The brute force approach to implementing the Gaussian filter.