d43682d51b
New features: * Bump mapping now works with SSS * Texture Blur factor for SSS, see the documentation for details: http://wiki.blender.org/index.php/Doc:2.6/Manual/Render/Cycles/Nodes/Shaders#Subsurface_Scattering Work in progress for feedback: Initial implementation of the "BSSRDF Importance Sampling" paper, which uses a different importance sampling method. It gives better quality results in many ways, with the availability of both Cubic and Gaussian falloff functions, but also tends to be more noisy when using the progressive integrator and does not give great results with some geometry. It works quite well for the non-progressive integrator and is often less noisy there. This code may still change a lot, so unless you're testing it may be best to stick to the Compatible falloff function. Skin test render and file that takes advantage of the gaussian falloff: http://www.pasteall.org/pic/show.php?id=57661 http://www.pasteall.org/pic/show.php?id=57662 http://www.pasteall.org/blend/23501
137 lines
3.8 KiB
C++
137 lines
3.8 KiB
C++
/*
|
|
* Copyright 2011, Blender Foundation.
|
|
*
|
|
* This program is free software; you can redistribute it and/or
|
|
* modify it under the terms of the GNU General Public License
|
|
* as published by the Free Software Foundation; either version 2
|
|
* of the License, or (at your option) any later version.
|
|
*
|
|
* This program is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
* GNU General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU General Public License
|
|
* along with this program; if not, write to the Free Software Foundation,
|
|
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
|
|
*/
|
|
|
|
#include "bssrdf.h"
|
|
|
|
#include "util_algorithm.h"
|
|
#include "util_math.h"
|
|
#include "util_types.h"
|
|
|
|
#include "kernel_types.h"
|
|
#include "kernel_montecarlo.h"
|
|
|
|
CCL_NAMESPACE_BEGIN
|
|
|
|
static float bssrdf_cubic(float ld, float r)
|
|
{
|
|
if(ld == 0.0f)
|
|
return (r == 0.0f)? 1.0f: 0.0f;
|
|
|
|
return powf(ld - min(r, ld), 3.0f) * 4.0f/powf(ld, 4.0f);
|
|
}
|
|
|
|
/* Cumulative density function utilities */
|
|
|
|
static float cdf_lookup_inverse(const vector<float>& table, float2 range, float x)
|
|
{
|
|
int index = upper_bound(table.begin(), table.end(), x) - table.begin();
|
|
|
|
if(index == 0)
|
|
return range[0];
|
|
else if(index == table.size())
|
|
return range[1];
|
|
else
|
|
index--;
|
|
|
|
float t = (x - table[index])/(table[index+1] - table[index]);
|
|
float y = ((index + t)/(table.size() - 1));
|
|
|
|
return y*(range[1] - range[0]) + range[0];
|
|
}
|
|
|
|
static void cdf_invert(vector<float>& to, float2 to_range, const vector<float>& from, float2 from_range)
|
|
{
|
|
float step = 1.0f/(float)(to.size() - 1);
|
|
|
|
for(int i = 0; i < to.size(); i++) {
|
|
float x = (i*step)*(from_range[1] - from_range[0]) + from_range[0];
|
|
to[i] = cdf_lookup_inverse(from, to_range, x);
|
|
}
|
|
}
|
|
|
|
/* BSSRDF */
|
|
|
|
static void bssrdf_lookup_table_create(float ld, vector<float>& sample_table, vector<float>& pdf_table)
|
|
{
|
|
const int size = BSSRDF_RADIUS_TABLE_SIZE;
|
|
vector<float> cdf(size);
|
|
vector<float> pdf(size);
|
|
float step = 1.0f/(float)(size - 1);
|
|
float max_radius = ld;
|
|
float pdf_sum = 0.0f;
|
|
|
|
/* compute the probability density function */
|
|
for(int i = 0; i < pdf.size(); i++) {
|
|
float x = (i*step)*max_radius;
|
|
pdf[i] = bssrdf_cubic(ld, x);
|
|
pdf_sum += pdf[i];
|
|
}
|
|
|
|
/* adjust for area covered by each distance */
|
|
for(int i = 0; i < pdf.size(); i++) {
|
|
float x = (i*step)*max_radius;
|
|
pdf[i] *= M_2PI_F*x;
|
|
}
|
|
|
|
/* normalize pdf, we multiply in reflectance later */
|
|
if(pdf_sum > 0.0f)
|
|
for(int i = 0; i < pdf.size(); i++)
|
|
pdf[i] /= pdf_sum;
|
|
|
|
/* sum to account for sampling which uses overlapping sphere */
|
|
for(int i = pdf.size() - 2; i >= 0; i--)
|
|
pdf[i] = pdf[i] + pdf[i+1];
|
|
|
|
/* compute the cumulative density function */
|
|
cdf[0] = 0.0f;
|
|
|
|
for(int i = 1; i < size; i++)
|
|
cdf[i] = cdf[i-1] + 0.5f*(pdf[i-1] + pdf[i])*step*max_radius;
|
|
|
|
/* invert cumulative density function for importance sampling */
|
|
float2 cdf_range = make_float2(0.0f, cdf[size - 1]);
|
|
float2 table_range = make_float2(0.0f, max_radius);
|
|
|
|
cdf_invert(sample_table, table_range, cdf, cdf_range);
|
|
|
|
/* copy pdf table */
|
|
for(int i = 0; i < pdf.size(); i++)
|
|
pdf_table[i] = pdf[i];
|
|
}
|
|
|
|
void bssrdf_table_build(vector<float>& table)
|
|
{
|
|
vector<float> sample_table(BSSRDF_RADIUS_TABLE_SIZE);
|
|
vector<float> pdf_table(BSSRDF_RADIUS_TABLE_SIZE);
|
|
|
|
table.resize(BSSRDF_LOOKUP_TABLE_SIZE);
|
|
|
|
/* create a 2D lookup table, for reflection x sample radius */
|
|
for(int i = 0; i < BSSRDF_REFL_TABLE_SIZE; i++) {
|
|
float radius = 1.0f;
|
|
|
|
bssrdf_lookup_table_create(radius, sample_table, pdf_table);
|
|
|
|
memcpy(&table[i*BSSRDF_RADIUS_TABLE_SIZE], &sample_table[0], BSSRDF_RADIUS_TABLE_SIZE*sizeof(float));
|
|
memcpy(&table[BSSRDF_PDF_TABLE_OFFSET + i*BSSRDF_RADIUS_TABLE_SIZE], &pdf_table[0], BSSRDF_RADIUS_TABLE_SIZE*sizeof(float));
|
|
}
|
|
}
|
|
|
|
CCL_NAMESPACE_END
|
|
|