griefith/test
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Nodes: Optimize Gabor noise with early exit

Browse Source 
Optimize the Gabor noise texture code with an early exit for points that
are further away from the kernel center. This was already done for the
kernel, but is now being done earlier before computing the weight, so
its computation is now skipped.

Thanks to Charlie Jolly for the suggestion.
This commit is contained in:
Omar Emara
2024-08-14 12:12:34 +03:00
parent 8ccf6036e6
commit 847a3a7ea3
4 changed files with 52 additions and 50 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
source/blender/gpu/shaders/material/gpu_shader_material_tex_gabor.glsl
View File

@@ -60,13 +60,7 @@
* normalization", to ensure a zero mean, which should help with normalization. */
vec2 compute_2d_gabor_kernel(vec2 position, float frequency, float orientation)
{
/* The kernel is windowed beyond the unit distance, so early exist with a zero for points that
* are further than a unit radius. */
float distance_squared = length_squared(position);
if (distance_squared >= 1.0) {
return vec2(0.0);
}
float hann_window = 0.5 + 0.5 * cos(M_PI * distance_squared);
float gaussian_envelop = exp(-M_PI * distance_squared);
float windowed_gaussian_envelope = gaussian_envelop * hann_window;
@@ -130,6 +124,12 @@ vec2 compute_2d_gabor_noise_cell(
vec2 kernel_center = hash_vec3_to_vec2(seed_for_kernel_center);
vec2 position_in_kernel_space = position - kernel_center;
/* The kernel is windowed beyond the unit distance, so early exit with a zero for points that
* are further than a unit radius. */
if (length_squared(position_in_kernel_space) >= 1.0) {
continue;
}
/* We either add or subtract the Gabor kernel based on a Bernoulli distribution of equal
* probability. */
float weight = hash_vec3_to_float(seed_for_weight) < 0.5 ? -1.0 : 1.0;
@@ -153,8 +153,10 @@ vec2 compute_2d_gabor_noise(vec2 coordinates,
for (int j = -1; j <= 1; j++) {
for (int i = -1; i <= 1; i++) {
vec2 cell_offset = vec2(i, j);
vec2 current_cell_position = cell_position + cell_offset;
vec2 position_in_cell_space = local_position - cell_offset;
sum += compute_2d_gabor_noise_cell(
current_cell_position, position_in_cell_space, frequency, isotropy, base_orientation);
}
@@ -169,13 +171,7 @@ vec2 compute_2d_gabor_noise(vec2 coordinates,
* vector, so we just need to scale it by the frequency value. */
vec2 compute_3d_gabor_kernel(vec3 position, float frequency, vec3 orientation)
{
/* The kernel is windowed beyond the unit distance, so early exist with a zero for points that
* are further than a unit radius. */
float distance_squared = length_squared(position);
if (distance_squared >= 1.0) {
return vec2(0.0);
}
float hann_window = 0.5 + 0.5 * cos(M_PI * distance_squared);
float gaussian_envelop = exp(-M_PI * distance_squared);
float windowed_gaussian_envelope = gaussian_envelop * hann_window;
@@ -240,6 +236,12 @@ vec2 compute_3d_gabor_noise_cell(
vec3 kernel_center = hash_vec4_to_vec3(seed_for_kernel_center);
vec3 position_in_kernel_space = position - kernel_center;
/* The kernel is windowed beyond the unit distance, so early exit with a zero for points that
* are further than a unit radius. */
if (length_squared(position_in_kernel_space) >= 1.0) {
continue;
}
/* We either add or subtract the Gabor kernel based on a Bernoulli distribution of equal
* probability. */
float weight = hash_vec4_to_float(seed_for_weight) < 0.5 ? -1.0 : 1.0;