typename Kernel::Model MaxConsensus
(
const Kernel &kernel,
const Scorer &scorer,
std::vector<uint32_t> *best_inliers = nullptr,
uint32_t max_iteration = 1024
)
{
const uint32_t min_samples = Kernel::MINIMUM_SAMPLES;
const uint32_t total_samples = kernel.NumSamples();
size_t best_num_inliers = 0;
typename Kernel::Model best_model;
// Test if we have sufficient points to for the kernel.
if (total_samples < min_samples) {
if (best_inliers) {
best_inliers->resize(0);
}
return best_model;
}
// In this robust estimator, the scorer always works on all the data points
// at once. So precompute the list ahead of time.
std::vector<uint32_t> all_samples(total_samples);
std::iota(all_samples.begin(), all_samples.end(), 0);
// Random number generator configuration
std::mt19937 random_generator(std::mt19937::default_seed);
std::vector<uint32_t> sample;
for (uint32_t iteration = 0; iteration < max_iteration; ++iteration) {
UniformSample(min_samples, random_generator, &all_samples, &sample);
std::vector<typename Kernel::Model> models;
kernel.Fit(sample, &models);
// Compute costs for each fit.
for (const auto& model_it : models) {
std::vector<uint32_t> inliers;
scorer.Score(kernel, model_it, all_samples, &inliers);
if (best_num_inliers < inliers.size()) {
best_num_inliers = inliers.size();
best_model = model_it;
if (best_inliers) {
best_inliers->swap(inliers);
}
}
}
}
return best_model;
}
typename Kernel::Model RANSAC(
const Kernel &kernel,
const Scorer &scorer,
std::vector<size_t> *best_inliers = nullptr ,
size_t *best_score = nullptr , // Found number of inliers
double outliers_probability = 1e-2)
{
assert(outliers_probability < 1.0);
assert(outliers_probability > 0.0);
size_t iteration = 0;
const size_t min_samples = Kernel::MINIMUM_SAMPLES;
const size_t total_samples = kernel.NumSamples();
size_t max_iterations = 100;
const size_t really_max_iterations = 4096;
size_t best_num_inliers = 0;
double best_inlier_ratio = 0.0;
typename Kernel::Model best_model;
// Test if we have sufficient points for the kernel.
if (total_samples < min_samples) {
if (best_inliers) {
best_inliers->resize(0);
}
return best_model;
}
// In this robust estimator, the scorer always works on all the data points
// at once. So precompute the list ahead of time [0,..,total_samples].
std::vector<size_t> all_samples(total_samples);
std::iota(all_samples.begin(), all_samples.end(), 0);
std::vector<size_t> sample;
for (iteration = 0;
iteration < max_iterations &&
iteration < really_max_iterations; ++iteration) {
UniformSample(min_samples, &all_samples, &sample);
std::vector<typename Kernel::Model> models;
kernel.Fit(sample, &models);
// Compute the inlier list for each fit.
for (size_t i = 0; i < models.size(); ++i) {
std::vector<size_t> inliers;
scorer.Score(kernel, models[i], all_samples, &inliers);
if (best_num_inliers < inliers.size()) {
best_num_inliers = inliers.size();
best_inlier_ratio = inliers.size() / double(total_samples);
best_model = models[i];
if (best_inliers) {
best_inliers->swap(inliers);
}
}
if (best_inlier_ratio) {
max_iterations = IterationsRequired(min_samples,
outliers_probability,
best_inlier_ratio);
}
}
}
if (best_score)
*best_score = best_num_inliers;
return best_model;
}
bool
OpenEXRInput::read_native_deep_tiles (int xbegin, int xend,
int ybegin, int yend,
int zbegin, int zend,
int chbegin, int chend,
DeepData &deepdata)
{
if (m_deep_tiled_input_part == NULL) {
error ("called OpenEXRInput::read_native_deep_tiles without an open file");
return false;
}
try {
const PartInfo &part (m_parts[m_subimage]);
size_t width = (xend - xbegin);
size_t npixels = width * (yend - ybegin) * (zend - zbegin);
chend = clamp (chend, chbegin+1, m_spec.nchannels);
int nchans = chend - chbegin;
// Set up the count and pointers arrays and the Imf framebuffer
std::vector<TypeDesc> channeltypes;
m_spec.get_channelformats (channeltypes);
deepdata.init (npixels, nchans,
array_view<const TypeDesc>(&channeltypes[chbegin], chend-chbegin),
spec().channelnames);
std::vector<unsigned int> all_samples (npixels);
std::vector<void*> pointerbuf (npixels * nchans);
Imf::DeepFrameBuffer frameBuffer;
Imf::Slice countslice (Imf::UINT,
(char *)(&all_samples[0]
- xbegin
- ybegin*width),
sizeof(unsigned int),
sizeof(unsigned int) * width);
frameBuffer.insertSampleCountSlice (countslice);
for (int c = chbegin; c < chend; ++c) {
Imf::DeepSlice slice (part.pixeltype[c],
(char *)(&pointerbuf[0]+(c-chbegin)
- xbegin*nchans
- ybegin*width*nchans),
sizeof(void*) * nchans, // xstride of pointer array
sizeof(void*) * nchans*width, // ystride of pointer array
deepdata.samplesize()); // stride of data sample
frameBuffer.insert (m_spec.channelnames[c].c_str(), slice);
}
m_deep_tiled_input_part->setFrameBuffer (frameBuffer);
int xtiles = round_to_multiple (xend-xbegin, m_spec.tile_width) / m_spec.tile_width;
int ytiles = round_to_multiple (yend-ybegin, m_spec.tile_height) / m_spec.tile_height;
int firstxtile = (xbegin - m_spec.x) / m_spec.tile_width;
int firstytile = (ybegin - m_spec.y) / m_spec.tile_height;
// Get the sample counts for each pixel and compute the total
// number of samples and resize the data area appropriately.
m_deep_tiled_input_part->readPixelSampleCounts (
firstxtile, firstxtile+xtiles-1,
firstytile, firstytile+ytiles-1);
deepdata.set_all_samples (all_samples);
deepdata.get_pointers (pointerbuf);
// Read the pixels
m_deep_tiled_input_part->readTiles (
firstxtile, firstxtile+xtiles-1,
firstytile, firstytile+ytiles-1,
m_miplevel, m_miplevel);
} catch (const std::exception &e) {
error ("Failed OpenEXR read: %s", e.what());
return false;
} catch (...) { // catch-all for edge cases or compiler bugs
error ("Failed OpenEXR read: unknown exception");
return false;
}
return true;
}
bool
OpenEXRInput::read_native_deep_scanlines (int ybegin, int yend, int z,
int chbegin, int chend,
DeepData &deepdata)
{
if (m_deep_scanline_input_part == NULL) {
error ("called OpenEXRInput::read_native_deep_scanlines without an open file");
return false;
}
#ifdef USE_OPENEXR_VERSION2
try {
const PartInfo &part (m_parts[m_subimage]);
size_t npixels = (yend - ybegin) * m_spec.width;
chend = clamp (chend, chbegin+1, m_spec.nchannels);
int nchans = chend - chbegin;
// Set up the count and pointers arrays and the Imf framebuffer
std::vector<TypeDesc> channeltypes;
m_spec.get_channelformats (channeltypes);
deepdata.init (npixels, nchans,
array_view<const TypeDesc>(&channeltypes[chbegin], chend-chbegin),
spec().channelnames);
std::vector<unsigned int> all_samples (npixels);
std::vector<void*> pointerbuf (npixels*nchans);
Imf::DeepFrameBuffer frameBuffer;
Imf::Slice countslice (Imf::UINT,
(char *)(&all_samples[0]
- m_spec.x
- ybegin*m_spec.width),
sizeof(unsigned int),
sizeof(unsigned int) * m_spec.width);
frameBuffer.insertSampleCountSlice (countslice);
for (int c = chbegin; c < chend; ++c) {
Imf::DeepSlice slice (part.pixeltype[c],
(char *)(&pointerbuf[0]+(c-chbegin)
- m_spec.x * nchans
- ybegin*m_spec.width*nchans),
sizeof(void*) * nchans, // xstride of pointer array
sizeof(void*) * nchans*m_spec.width, // ystride of pointer array
deepdata.samplesize()); // stride of data sample
frameBuffer.insert (m_spec.channelnames[c].c_str(), slice);
}
m_deep_scanline_input_part->setFrameBuffer (frameBuffer);
// Get the sample counts for each pixel and compute the total
// number of samples and resize the data area appropriately.
m_deep_scanline_input_part->readPixelSampleCounts (ybegin, yend-1);
deepdata.set_all_samples (all_samples);
deepdata.get_pointers (pointerbuf);
// Read the pixels
m_deep_scanline_input_part->readPixels (ybegin, yend-1);
// deepdata.import_chansamp (pointerbuf);
} catch (const std::exception &e) {
error ("Failed OpenEXR read: %s", e.what());
return false;
} catch (...) { // catch-all for edge cases or compiler bugs
error ("Failed OpenEXR read: unknown exception");
return false;
}
return true;
#else
return false;
#endif
}
