void copyTexGPUMem(const gpu::Texture& texture, GLenum texTarget, GLuint srcId, GLuint destId, uint16_t numMips, uint16_t srcMipOffset, uint16_t destMipOffset, uint16_t populatedMips) {
for (uint16_t mip = populatedMips; mip < numMips; ++mip) {
auto mipDimensions = texture.evalMipDimensions(mip);
uint16_t targetMip = mip - destMipOffset;
uint16_t sourceMip = mip - srcMipOffset;
auto faces = GLTexture::getFaceCount(texTarget);
for (uint8_t face = 0; face < faces; ++face) {
glCopyImageSubData(
srcId, texTarget, sourceMip, 0, 0, face,
destId, texTarget, targetMip, 0, 0, face,
mipDimensions.x, mipDimensions.y, 1
);
(void)CHECK_GL_ERROR();
}
}
}
bool sphericalHarmonicsFromTexture(const gpu::Texture& cubeTexture, std::vector<glm::vec3> & output, const uint order) {
int width = cubeTexture.getWidth();
if(width != cubeTexture.getHeight()) {
return false;
}
PROFILE_RANGE(render_gpu, "sphericalHarmonicsFromTexture");
auto mipFormat = cubeTexture.getStoredMipFormat();
std::function<glm::vec3(uint32)> unpackFunc;
switch (mipFormat.getSemantic()) {
case gpu::R11G11B10:
unpackFunc = glm::unpackF2x11_1x10;
break;
case gpu::RGB9E5:
unpackFunc = glm::unpackF3x9_E1x5;
break;
default:
assert(false);
break;
}
const uint sqOrder = order*order;
// allocate memory for calculations
output.resize(sqOrder);
std::vector<float> resultR(sqOrder);
std::vector<float> resultG(sqOrder);
std::vector<float> resultB(sqOrder);
// initialize values
float fWt = 0.0f;
for(uint i=0; i < sqOrder; i++) {
output[i] = glm::vec3(0.0f);
resultR[i] = 0.0f;
resultG[i] = 0;
resultB[i] = 0;
}
std::vector<float> shBuff(sqOrder);
std::vector<float> shBuffB(sqOrder);
// We trade accuracy for speed by breaking the image into 32x32 parts
// and approximating the distance for all the pixels in each part to be
// the distance to the part's center.
int numDivisionsPerSide = 32;
if (width < numDivisionsPerSide) {
numDivisionsPerSide = width;
}
int stride = width / numDivisionsPerSide;
int halfStride = stride / 2;
// for each face of cube texture
for(int face=0; face < gpu::Texture::NUM_CUBE_FACES; face++) {
PROFILE_RANGE(render_gpu, "ProcessFace");
auto data = reinterpret_cast<const uint32*>( cubeTexture.accessStoredMipFace(0, face)->readData() );
if (data == nullptr) {
continue;
}
// step between two texels for range [0, 1]
float invWidth = 1.0f / float(width);
// initial negative bound for range [-1, 1]
float negativeBound = -1.0f + invWidth;
// step between two texels for range [-1, 1]
float invWidthBy2 = 2.0f / float(width);
for(int y=halfStride; y < width-halfStride; y += stride) {
// texture coordinate V in range [-1 to 1]
const float fV = negativeBound + float(y) * invWidthBy2;
for(int x=halfStride; x < width - halfStride; x += stride) {
// texture coordinate U in range [-1 to 1]
const float fU = negativeBound + float(x) * invWidthBy2;
// determine direction from center of cube texture to current texel
glm::vec3 dir;
switch(face) {
case gpu::Texture::CUBE_FACE_RIGHT_POS_X: {
dir.x = 1.0f;
dir.y = 1.0f - (invWidthBy2 * float(y) + invWidth);
dir.z = 1.0f - (invWidthBy2 * float(x) + invWidth);
dir = -dir;
break;
}
case gpu::Texture::CUBE_FACE_LEFT_NEG_X: {
dir.x = -1.0f;
dir.y = 1.0f - (invWidthBy2 * float(y) + invWidth);
dir.z = -1.0f + (invWidthBy2 * float(x) + invWidth);
dir = -dir;
break;
}
case gpu::Texture::CUBE_FACE_TOP_POS_Y: {
dir.x = - 1.0f + (invWidthBy2 * float(x) + invWidth);
dir.y = 1.0f;
dir.z = - 1.0f + (invWidthBy2 * float(y) + invWidth);
dir = -dir;
break;
}
case gpu::Texture::CUBE_FACE_BOTTOM_NEG_Y: {
dir.x = - 1.0f + (invWidthBy2 * float(x) + invWidth);
dir.y = - 1.0f;
dir.z = 1.0f - (invWidthBy2 * float(y) + invWidth);
dir = -dir;
break;
}
case gpu::Texture::CUBE_FACE_BACK_POS_Z: {
dir.x = - 1.0f + (invWidthBy2 * float(x) + invWidth);
dir.y = 1.0f - (invWidthBy2 * float(y) + invWidth);
dir.z = 1.0f;
break;
}
case gpu::Texture::CUBE_FACE_FRONT_NEG_Z: {
dir.x = 1.0f - (invWidthBy2 * float(x) + invWidth);
dir.y = 1.0f - (invWidthBy2 * float(y) + invWidth);
dir.z = - 1.0f;
break;
}
default:
return false;
}
// normalize direction
dir = glm::normalize(dir);
// scale factor depending on distance from center of the face
const float fDiffSolid = 4.0f / ((1.0f + fU*fU + fV*fV) *
sqrtf(1.0f + fU*fU + fV*fV));
fWt += fDiffSolid;
// calculate coefficients of spherical harmonics for current direction
sphericalHarmonicsEvaluateDirection(shBuff.data(), order, dir);
// index of texel in texture
// get color from texture
glm::vec3 color{ 0.0f, 0.0f, 0.0f };
for (int i = 0; i < stride; ++i) {
for (int j = 0; j < stride; ++j) {
int k = (int)(x + i - halfStride + (y + j - halfStride) * width);
color += unpackFunc(data[k]);
}
}
// scale color and add to previously accumulated coefficients
// red
sphericalHarmonicsScale(shBuffB.data(), order, shBuff.data(), color.r * fDiffSolid);
sphericalHarmonicsAdd(resultR.data(), order, resultR.data(), shBuffB.data());
// green
sphericalHarmonicsScale(shBuffB.data(), order, shBuff.data(), color.g * fDiffSolid);
sphericalHarmonicsAdd(resultG.data(), order, resultG.data(), shBuffB.data());
// blue
sphericalHarmonicsScale(shBuffB.data(), order, shBuff.data(), color.b * fDiffSolid);
sphericalHarmonicsAdd(resultB.data(), order, resultB.data(), shBuffB.data());
}
}
}
// final scale for coefficients
const float fNormProj = (4.0f * glm::pi<float>()) / (fWt * (float)(stride * stride));
sphericalHarmonicsScale(resultR.data(), order, resultR.data(), fNormProj);
sphericalHarmonicsScale(resultG.data(), order, resultG.data(), fNormProj);
sphericalHarmonicsScale(resultB.data(), order, resultB.data(), fNormProj);
// save result
for(uint i=0; i < sqOrder; i++) {
output[i] = glm::vec3(resultR[i], resultG[i], resultB[i]);
}
return true;
}
bool sphericalHarmonicsFromTexture(const gpu::Texture& cubeTexture, std::vector<glm::vec3> & output, const uint order) {
const uint sqOrder = order*order;
// allocate memory for calculations
output.resize(sqOrder);
std::vector<float> resultR(sqOrder);
std::vector<float> resultG(sqOrder);
std::vector<float> resultB(sqOrder);
int width, height;
// initialize values
float fWt = 0.0f;
for(uint i=0; i < sqOrder; i++) {
output[i] = glm::vec3(0.0f);
resultR[i] = 0.0f;
resultG[i] = 0;
resultB[i] = 0;
}
std::vector<float> shBuff(sqOrder);
std::vector<float> shBuffB(sqOrder);
// get width and height
width = height = cubeTexture.getWidth();
if(width != height) {
return false;
}
const float UCHAR_TO_FLOAT = 1.0f / float(std::numeric_limits<unsigned char>::max());
// for each face of cube texture
for(int face=0; face < gpu::Texture::NUM_CUBE_FACES; face++) {
auto numComponents = cubeTexture.accessStoredMipFace(0,face)->_format.getDimensionCount();
auto data = cubeTexture.accessStoredMipFace(0,face)->_sysmem.readData();
if (data == nullptr) {
continue;
}
// step between two texels for range [0, 1]
float invWidth = 1.0f / float(width);
// initial negative bound for range [-1, 1]
float negativeBound = -1.0f + invWidth;
// step between two texels for range [-1, 1]
float invWidthBy2 = 2.0f / float(width);
for(int y=0; y < width; y++) {
// texture coordinate V in range [-1 to 1]
const float fV = negativeBound + float(y) * invWidthBy2;
for(int x=0; x < width; x++) {
// texture coordinate U in range [-1 to 1]
const float fU = negativeBound + float(x) * invWidthBy2;
// determine direction from center of cube texture to current texel
glm::vec3 dir;
switch(face) {
case gpu::Texture::CUBE_FACE_RIGHT_POS_X: {
dir.x = 1.0f;
dir.y = 1.0f - (invWidthBy2 * float(y) + invWidth);
dir.z = 1.0f - (invWidthBy2 * float(x) + invWidth);
dir = -dir;
break;
}
case gpu::Texture::CUBE_FACE_LEFT_NEG_X: {
dir.x = -1.0f;
dir.y = 1.0f - (invWidthBy2 * float(y) + invWidth);
dir.z = -1.0f + (invWidthBy2 * float(x) + invWidth);
dir = -dir;
break;
}
case gpu::Texture::CUBE_FACE_TOP_POS_Y: {
dir.x = - 1.0f + (invWidthBy2 * float(x) + invWidth);
dir.y = 1.0f;
dir.z = - 1.0f + (invWidthBy2 * float(y) + invWidth);
dir = -dir;
break;
}
case gpu::Texture::CUBE_FACE_BOTTOM_NEG_Y: {
dir.x = - 1.0f + (invWidthBy2 * float(x) + invWidth);
dir.y = - 1.0f;
dir.z = 1.0f - (invWidthBy2 * float(y) + invWidth);
dir = -dir;
break;
}
case gpu::Texture::CUBE_FACE_BACK_POS_Z: {
dir.x = - 1.0f + (invWidthBy2 * float(x) + invWidth);
dir.y = 1.0f - (invWidthBy2 * float(y) + invWidth);
dir.z = 1.0f;
break;
}
case gpu::Texture::CUBE_FACE_FRONT_NEG_Z: {
dir.x = 1.0f - (invWidthBy2 * float(x) + invWidth);
dir.y = 1.0f - (invWidthBy2 * float(y) + invWidth);
dir.z = - 1.0f;
break;
}
default:
return false;
}
// normalize direction
dir = glm::normalize(dir);
// scale factor depending on distance from center of the face
const float fDiffSolid = 4.0f / ((1.0f + fU*fU + fV*fV) *
sqrtf(1.0f + fU*fU + fV*fV));
fWt += fDiffSolid;
// calculate coefficients of spherical harmonics for current direction
sphericalHarmonicsEvaluateDirection(shBuff.data(), order, dir);
// index of texel in texture
uint pixOffsetIndex = (x + y * width) * numComponents;
// get color from texture and map to range [0, 1]
glm::vec3 clr(float(data[pixOffsetIndex]) * UCHAR_TO_FLOAT,
float(data[pixOffsetIndex+1]) * UCHAR_TO_FLOAT,
float(data[pixOffsetIndex+2]) * UCHAR_TO_FLOAT);
// Gamma correct
clr = sRGBToLinear(clr);
// scale color and add to previously accumulated coefficients
sphericalHarmonicsScale(shBuffB.data(), order,
shBuff.data(), clr.r * fDiffSolid);
sphericalHarmonicsAdd(resultR.data(), order,
resultR.data(), shBuffB.data());
sphericalHarmonicsScale(shBuffB.data(), order,
shBuff.data(), clr.g * fDiffSolid);
sphericalHarmonicsAdd(resultG.data(), order,
resultG.data(), shBuffB.data());
sphericalHarmonicsScale(shBuffB.data(), order,
shBuff.data(), clr.b * fDiffSolid);
sphericalHarmonicsAdd(resultB.data(), order,
resultB.data(), shBuffB.data());
}
}
}
// final scale for coefficients
const float fNormProj = (4.0f * glm::pi<float>()) / fWt;
sphericalHarmonicsScale(resultR.data(), order, resultR.data(), fNormProj);
sphericalHarmonicsScale(resultG.data(), order, resultG.data(), fNormProj);
sphericalHarmonicsScale(resultB.data(), order, resultB.data(), fNormProj);
// save result
for(uint i=0; i < sqOrder; i++) {
// gamma Correct
// output[i] = linearTosRGB(glm::vec3(resultR[i], resultG[i], resultB[i]));
output[i] = glm::vec3(resultR[i], resultG[i], resultB[i]);
}
return true;
}