ShapeContext3DEstimation implements the 3D shape context descriptor as described in: More...
#include <3dsc.h>
Inheritance diagram for pcl::ShapeContext3DEstimation< PointInT, PointNT, Eigen::MatrixXf >:
Private Member Functions | |
| void | compute (pcl::PointCloud< pcl::SHOT > &) |
| Make the compute (&PointCloudOut); inaccessible from outside the class. | |
| void | computeFeatureEigen (pcl::PointCloud< Eigen::MatrixXf > &output) |
| Estimate the actual feature. | |
Detailed Description
template<typename PointInT, typename PointNT>
class pcl::ShapeContext3DEstimation< PointInT, PointNT, Eigen::MatrixXf >
ShapeContext3DEstimation implements the 3D shape context descriptor as described in:
- Andrea Frome, Daniel Huber, Ravi Kolluri and Thomas Bülow, Jitendra Malik Recognizing Objects in Range Data Using Regional Point Descriptors, In proceedings of the 8th European Conference on Computer Vision (ECCV), Prague, May 11-14, 2004
The 3DSC computed feature has the following structure
- rf float[9] = x_axis | y_axis | normal and represents the local frame
- desc std::vector<float> which size is determined by the number of bins radius_bins_ + elevation_bins_ + azimuth_bins_. If shift is required then the computed descriptor will be shift along the azimuthal direction.
- Attention:
- The convention for a 3D shape context descriptor is:
- if a query point's nearest neighbors cannot be estimated, the feature descriptor will be set to NaN (not a number), and the RF to 0
- it is impossible to estimate a 3D shape context descriptor for a point that doesn't have finite 3D coordinates. Therefore, any point that contains NaN data on x, y, or z, will have its boundary feature property set to NaN.
Member Function Documentation
template<typename PointInT , typename PointNT >
| void pcl::ShapeContext3DEstimation< PointInT, PointNT, Eigen::MatrixXf >::compute | ( | pcl::PointCloud< pcl::SHOT > & | ) | [inline, private] |
template<typename PointInT , typename PointNT >
| void pcl::ShapeContext3DEstimation< PointInT, PointNT, Eigen::MatrixXf >::computeFeatureEigen | ( | pcl::PointCloud< Eigen::MatrixXf > & | output | ) | [private, virtual] |
Estimate the actual feature.
- Parameters:
-
[out] output the resultant feature
Reimplemented from pcl::ShapeContext3DEstimation< PointInT, PointNT, pcl::SHOT >.
The documentation for this class was generated from the following files: