LeastMedianSquares represents an implementation of the LMedS (Least Median of Squares) algorithm. LMedS is a RANSAC-like model-fitting algorithm that can tolerate up to 50% outliers without requiring thresholds to be set. See Andrea Fusiello's "Elements of Geometric Computer Vision" (http://homepages.inf.ed.ac.uk/rbf/CVonline/LOCAL_COPIES/FUSIELLO4/tutorial.html#x1-520007) for more details. More...
#include <lmeds.h>
Public Member Functions | |
bool | computeModel (int debug_verbosity_level=0) |
Compute the actual model and find the inliers. | |
LeastMedianSquares (const SampleConsensusModelPtr &model, double threshold) | |
LMedS (Least Median of Squares) main constructor. | |
LeastMedianSquares (const SampleConsensusModelPtr &model) | |
LMedS (Least Median of Squares) main constructor. | |
Private Types | |
typedef SampleConsensusModel < PointT >::Ptr | SampleConsensusModelPtr |
LeastMedianSquares represents an implementation of the LMedS (Least Median of Squares) algorithm. LMedS is a RANSAC-like model-fitting algorithm that can tolerate up to 50% outliers without requiring thresholds to be set. See Andrea Fusiello's "Elements of Geometric Computer Vision" (http://homepages.inf.ed.ac.uk/rbf/CVonline/LOCAL_COPIES/FUSIELLO4/tutorial.html#x1-520007) for more details.
Definition at line 54 of file lmeds.h.
typedef SampleConsensusModel<PointT>::Ptr pcl::LeastMedianSquares< PointT >::SampleConsensusModelPtr [private] |
Reimplemented from pcl::SampleConsensus< PointT >.
pcl::LeastMedianSquares< PointT >::LeastMedianSquares | ( | const SampleConsensusModelPtr & | model | ) | [inline] |
pcl::LeastMedianSquares< PointT >::LeastMedianSquares | ( | const SampleConsensusModelPtr & | model, | |
double | threshold | |||
) | [inline] |
bool pcl::LeastMedianSquares< PointT >::computeModel | ( | int | debug_verbosity_level = 0 |
) | [inline, virtual] |
Compute the actual model and find the inliers.
debug_verbosity_level | enable/disable on-screen debug information and set the verbosity level |
Implements pcl::SampleConsensus< PointT >.