33     std::vector<bool>& alreadySelectedOther
    34 #ifdef AVOID_MULTIPLE_CORRESPONDENCES
    36     const std::vector<std::vector<int>>& listDuplicatedLandmarksThis
    40     ASSERTDEB_(
c.this_idx < alreadySelectedThis.size());
    41     ASSERTDEB_(
c.other_idx < alreadySelectedOther.size());
    43 #ifndef AVOID_MULTIPLE_CORRESPONDENCES    44     alreadySelectedThis[
c.this_idx] = 
true;
    45     alreadySelectedOther[
c.other_idx] = 
true;
    48              listDuplicatedLandmarksThis[
c.this_idx].begin();
    49          it1 != listDuplicatedLandmarksThis[
c.this_idx].end(); it1++)
    50         alreadySelectedThis[*it1] = 
true;
    52              listDuplicatedLandmarksOther[
c.other_idx].begin();
    53          it2 != listDuplicatedLandmarksOther[
c.other_idx].end(); it2++)
    54         alreadySelectedOther[*it2] = 
true;
    89     const size_t nCorrs = in_correspondences.size();
    92     const double MAX_RMSE_TO_END =
    99     if (nCorrs < 
params.ransac_minSetSize)
   102         results.transformation.clear();
   108     timlog.enter(
"ransac.find_max*");
   111     unsigned int maxThis = 0, maxOther = 0;
   113          matchIt != in_correspondences.end(); ++matchIt)
   115         maxThis = max(maxThis, matchIt->this_idx);
   116         maxOther = max(maxOther, matchIt->other_idx);
   119     timlog.leave(
"ransac.find_max*");
   123     timlog.enter(
"ransac.count_unique_corrs");
   127     std::vector<bool> hasCorrThis(maxThis + 1, 
false);
   128     std::vector<bool> hasCorrOther(maxOther + 1, 
false);
   129     unsigned int howManyDifCorrs = 0;
   131          matchIt != in_correspondences.end(); ++matchIt)
   133         if (!hasCorrThis[matchIt->this_idx] &&
   134             !hasCorrOther[matchIt->other_idx])
   136             hasCorrThis[matchIt->this_idx] = 
true;
   137             hasCorrOther[matchIt->other_idx] = 
true;
   142     timlog.leave(
"ransac.count_unique_corrs");
   146     results.transformation.clear();
   150     if (howManyDifCorrs < 
params.ransac_minSetSize)
   153         results.transformation.clear();
   158 #ifdef AVOID_MULTIPLE_CORRESPONDENCES   165     std::vector<std::vector<int>> listDuplicatedLandmarksThis(maxThis + 1);
   167     for (k = 0; k < nCorrs - 1; k++)
   169         std::vector<int> duplis;
   170         for (
unsigned j = k; j < nCorrs - 1; j++)
   172             if (in_correspondences[k].this_x == in_correspondences[j].this_x &&
   173                 in_correspondences[k].this_y == in_correspondences[j].this_y &&
   174                 in_correspondences[k].this_z == in_correspondences[j].this_z)
   175                 duplis.push_back(in_correspondences[j].this_idx);
   177         listDuplicatedLandmarksThis[in_correspondences[k].this_idx] = duplis;
   180     std::vector<std::vector<int>> listDuplicatedLandmarksOther(maxOther + 1);
   181     for (k = 0; k < nCorrs - 1; k++)
   183         std::vector<int> duplis;
   184         for (
unsigned j = k; j < nCorrs - 1; j++)
   186             if (in_correspondences[k].other_x ==
   187                     in_correspondences[j].other_x &&
   188                 in_correspondences[k].other_y ==
   189                     in_correspondences[j].other_y &&
   190                 in_correspondences[k].other_z == in_correspondences[j].other_z)
   191                 duplis.push_back(in_correspondences[j].other_idx);
   193         listDuplicatedLandmarksOther[in_correspondences[k].other_idx] = duplis;
   197     std::deque<TMatchingPairList> alreadyAddedSubSets;
   202     const double ransac_consistency_test_chi2_quantile = 0.99;
   203     const double chi2_thres_dim1 =
   209     size_t largest_consensus_yet = 0;  
   210     double largestSubSet_RMSE = std::numeric_limits<double>::max();
   213     const bool use_dynamic_iter_number = 
results.ransac_iters == 0;
   214     if (use_dynamic_iter_number)
   217             params.probability_find_good_model > 0 &&
   218             params.probability_find_good_model < 1);
   224     std::vector<bool> alreadySelectedThis, alreadySelectedOther;
   226     if (!
params.ransac_algorithmForLandmarks)
   228         alreadySelectedThis.assign(maxThis + 1, 
false);
   229         alreadySelectedOther.assign(maxOther + 1, 
false);
   235     std::vector<size_t> corrsIdxs(nCorrs), corrsIdxsPermutation;
   236     for (
size_t i = 0; i < nCorrs; i++) corrsIdxs[i] = i;
   239     for (iter_idx = 0; iter_idx < 
results.ransac_iters;
   243         CTimeLoggerEntry tle(timlog, 
"ransac.iter");
   247         timlog.enter(
"ransac.permute");
   252         timlog.leave(
"ransac.permute");
   258         if (
params.ransac_algorithmForLandmarks)
   261             timlog.enter(
"ransac.reset_selection_marks");
   263             alreadySelectedThis.assign(maxThis + 1, 
false);
   264             alreadySelectedOther.assign(maxOther + 1, 
false);
   266             timlog.leave(
"ransac.reset_selection_marks");
   279         timlog.enter(
"ransac.inner_loops");
   281         for (
unsigned int j = 0;
   282              j < nCorrs && subSet.size() < 
params.ransac_maxSetSize; j++)
   284             const size_t idx = corrsIdxsPermutation[j];
   289             if (alreadySelectedThis[corr_j.
this_idx] ||
   294             if (
params.user_individual_compat_callback)
   299                 if (!
params.user_individual_compat_callback(pm))
   303             if (subSet.size() < 2)
   309                 subSet.push_back(corr_j);
   310                 markAsPicked(corr_j, alreadySelectedThis, alreadySelectedOther);
   312                 if (subSet.size() == 2)
   320                     const double corrs_dist1 =
   322                             subSet[0].this_x, subSet[0].this_y,
   323                             subSet[1].this_x, subSet[1].this_y);
   325                     const double corrs_dist2 =
   327                             subSet[0].other_x, subSet[0].other_y,
   328                             subSet[1].other_x, subSet[1].other_y);
   332                     const double corrs_dist_chi2 =
   337                     bool is_acceptable = (corrs_dist_chi2 < chi2_thres_dim1);
   350                             (referenceEstimation.
cov(2, 2) <
   358                         subSet.erase(subSet.begin() + (subSet.size() - 1));
   364                             corr_j, alreadySelectedThis, alreadySelectedOther);
   371                 timlog.enter(
"ransac.test_consistency");
   391                 const bool passTest =
   392                     maha_dist < 
params.ransac_mahalanobisDistanceThreshold;
   397                     subSet.push_back(corr_j);
   399                         corr_j, alreadySelectedThis, alreadySelectedOther);
   404                 timlog.leave(
"ransac.test_consistency");
   410         timlog.leave(
"ransac.inner_loops");
   413         const bool has_to_eval_RMSE =
   414             (subSet.size() >= 
params.ransac_minSetSize);
   418         double this_subset_RMSE = 0;
   419         if (has_to_eval_RMSE)
   422             CTimeLoggerEntry tle(timlog, 
"ransac.comp_rmse");
   430             for (
size_t k = 0; k < subSet.size(); k++)
   434                     subSet[k].other_x, subSet[k].other_y, gx, gy);
   437                     mrpt::math::distanceSqrBetweenPoints<double>(
   438                         subSet[k].this_x, subSet[k].this_y, gx, gy);
   440             this_subset_RMSE /= std::max(static_cast<size_t>(1), subSet.size());
   444             this_subset_RMSE = std::numeric_limits<double>::max();
   450         if (subSet.size() >= 
params.ransac_minSetSize)
   459             if (!
params.ransac_fuseByCorrsMatch)
   463                 for (
size_t i = 0; i < 
results.transformation.size(); i++)
   466                         results.transformation.get(i).mean.distanceTo(
   467                             referenceEstimation.
mean);
   468                     double diffPhi = fabs(
   470                             results.transformation.get(i).mean.phi() -
   471                             referenceEstimation.
mean.
phi()));
   472                     if (diffXY < 
params.ransac_fuseMaxDiffXY &&
   473                         diffPhi < 
params.ransac_fuseMaxDiffPhi)
   491                 for (
size_t i = 0; i < alreadyAddedSubSets.size(); i++)
   493                     if (subSet == alreadyAddedSubSets[i])
   501             if (indexFound != -1)
   504                 if (
params.ransac_algorithmForLandmarks)
   505                     results.transformation.get(indexFound).log_w = log(
   506                         1 + exp(
results.transformation.get(indexFound).log_w));
   508                     results.transformation.get(indexFound).log_w =
   510                             exp(
results.transformation.get(indexFound).log_w));
   515                 alreadyAddedSubSets.push_back(subSet);
   518                 if (
params.ransac_algorithmForLandmarks)
   519                     newSOGMode.
log_w = 0;  
   521                     newSOGMode.
log_w = log(static_cast<double>(subSet.size()));
   523                 newSOGMode.
mean = referenceEstimation.
mean;
   524                 newSOGMode.
cov = referenceEstimation.
cov;
   527                 results.transformation.push_back(newSOGMode);
   531         const size_t ninliers = subSet.size();
   532         if (largest_consensus_yet < ninliers)
   534             largest_consensus_yet = ninliers;
   537             if (use_dynamic_iter_number)
   542                 const double fracinliers =
   544                     static_cast<double>(howManyDifCorrs);  
   546                     1 - pow(fracinliers, static_cast<double>(
   549                 pNoOutliers = std::max(
   550                     std::numeric_limits<double>::epsilon(),
   553                     1.0 - std::numeric_limits<double>::epsilon(),
   557                     log(1 - 
params.probability_find_good_model) /
   560                 results.ransac_iters = std::max(
   564                     cout << 
"[tfest::RANSAC] Iter #" << iter_idx
   565                          << 
":est. # iters=" << 
results.ransac_iters
   566                          << 
" pNoOutliers=" << pNoOutliers
   567                          << 
" #inliers: " << ninliers << endl;
   572         if (subSet.size() >= 
params.ransac_minSetSize &&
   573             this_subset_RMSE < largestSubSet_RMSE)
   576                 cout << 
"[tfest::RANSAC] Iter #" << iter_idx
   577                      << 
" Better subset: " << subSet.size()
   578                      << 
" inliers, RMSE=" << this_subset_RMSE << endl;
   580             results.largestSubSet = subSet;
   581             largestSubSet_RMSE = this_subset_RMSE;
   585         if (subSet.size() >= 
params.ransac_minSetSize &&
   586             this_subset_RMSE < MAX_RMSE_TO_END)
   592         timlog.leave(
"ransac.iter");
   597         cout << 
"[tfest::RANSAC] Finished after " << iter_idx
   601     results.transformation.normalizeWeights();
   606         printf(
"nCorrs=%u\n", static_cast<unsigned int>(nCorrs));
   607         printf(
"Saving '_debug_in_correspondences.txt'...");
   608         in_correspondences.
dumpToFile(
"_debug_in_correspondences.txt");
   609         printf(
"Ok\n"); printf(
"Saving '_debug_results.transformation.txt'...");
   610         results.transformation.saveToTextFile(
   611             "_debug_results.transformation.txt");
 A namespace of pseudo-random numbers generators of diferent distributions. 
void permuteVector(const VEC &in_vector, VEC &out_result)
Returns a random permutation of a vector: all the elements of the input vector are in the output but ...
CPose2D mean
The mean value. 
const float normalizationStd
The struct for each mode: 
A gaussian distribution for 2D points. 
void dumpToFile(const std::string &fileName) const
Saves the correspondences to a text file. 
void markAsPicked(const TMatchingPair &c, std::vector< bool > &alreadySelectedThis, std::vector< bool > &alreadySelectedOther)
bool se2_l2(const mrpt::tfest::TMatchingPairList &in_correspondences, mrpt::math::TPose2D &out_transformation, mrpt::math::CMatrixDouble33 *out_estimateCovariance=nullptr)
Least-squares (L2 norm) solution to finding the optimal SE(2) (x,y,yaw) between two reference frames...
Parameters for se2_l2_robust(). 
double DEG2RAD(const double x)
Degrees to radians. 
void composePoint(const mrpt::math::TPoint2D &l, CPoint2DPDFGaussian &g) const
Returns the PDF of the 2D point  with "q"=this pose and "l" a point without uncertainty. 
#define MRPT_END_WITH_CLEAN_UP(stuff)
void composePoint(double lx, double ly, double &gx, double &gy) const
An alternative, slightly more efficient way of doing  with G and L being 2D points and P this 2D pose...
mrpt::math::CMatrixDouble33 cov
The 3x3 covariance matrix. 
T square(const T x)
Inline function for the square of a number. 
#define ASSERT_(f)
Defines an assertion mechanism. 
This base provides a set of functions for maths stuff. 
bool se2_l2_robust(const mrpt::tfest::TMatchingPairList &in_correspondences, const double in_normalizationStd, const TSE2RobustParams &in_ransac_params, TSE2RobustResult &out_results)
Robust least-squares (L2 norm) solution to finding the optimal SE(2) (x,y,yaw) between two reference ...
map< string, CVectorDouble > results
Declares a class that represents a Probability Density function (PDF) of a 2D pose ...
mrpt::system::CTimeLogger CTimeLogger
mrpt::math::CMatrixDouble33 cov
double mahalanobisDistanceToPoint(const double x, const double y) const
Returns the Mahalanobis distance from this PDF to some point. 
T wrapToPi(T a)
Modifies the given angle to translate it into the ]-pi,pi] range. 
Classes for 2D/3D geometry representation, both of single values and probability density distribution...
A structure for holding correspondences between two sets of points or points-like entities in 2D or 3...
This is the global namespace for all Mobile Robot Programming Toolkit (MRPT) libraries. 
double chi2inv(double P, unsigned int dim=1)
The "quantile" of the Chi-Square distribution, for dimension "dim" and probability 0<P<1 (the inverse...
#define ASSERTDEB_(f)
Defines an assertion mechanism - only when compiled in debug. 
const double & phi() const
Get the phi angle of the 2D pose (in radians) 
T distanceBetweenPoints(const T x1, const T y1, const T x2, const T y2)
Returns the distance between 2 points in 2D. 
CRandomGenerator & getRandomGenerator()
A static instance of a CRandomGenerator class, for use in single-thread applications. 
double log_w
The log-weight. 
Functions for estimating the optimal transformation between two frames of references given measuremen...
GLenum const GLfloat * params
const Scalar * const_iterator
For each individual-compatibility (IC) test, the indices of the candidate match between elements in b...
Output placeholder for se2_l2_robust()