curvatureEstimation.h

/*
 This Source Code Form is subject to the terms of the Mozilla Public
 License, v. 2.0. If a copy of the MPL was not distributed with this
 file, You can obtain one at http://mozilla.org/MPL/2.0/.
*/
 
#pragma once
 
#include "./defines.h"
 
namespace Ponca
{
    template < class DataPoint, class _NFilter, int DiffType, typename T>
    class NormalDerivativesCurvatureEstimator : public T
    {
    PONCA_FITTING_DECLARE_DEFAULT_TYPES
    PONCA_FITTING_DECLARE_MATRIX_TYPE
 
    protected:
        enum
        {
            Check = Base::PROVIDES_NORMAL_DERIVATIVE && Base::PROVIDES_PRINCIPAL_CURVATURES
        };
 
    private:
        typedef Eigen::Matrix<Scalar,3,2> Mat32; 
        typedef Eigen::Matrix<Scalar,2,2> Mat22; 
    public:
        PONCA_EXPLICIT_CAST_OPERATORS_DER(NormalDerivativesCurvatureEstimator,normalDerivativesCurvatureEstimator)
        PONCA_FITTING_DECLARE_FINALIZE
 
    private:
        PONCA_MULTIARCH [[nodiscard]] inline FIT_RESULT computeCurvature(bool useNormal = false);
 
    protected:
        PONCA_MULTIARCH [[nodiscard]] inline Mat32 tangentPlane(bool useNormal = false) const;
    };
 
 
template < class DataPoint, class _NFilter, int DiffType, typename T>
class NormalCovarianceCurvatureEstimator : public T
{
PONCA_FITTING_DECLARE_DEFAULT_TYPES
PONCA_FITTING_DECLARE_MATRIX_TYPE
PONCA_FITTING_DECLARE_DEFAULT_DER_TYPES
 
protected:
    enum
    {
        Check = Base::PROVIDES_PRINCIPAL_CURVATURES
    };
 
    //TODO(thib) check the curvature values that might be wrong
    //TODO(thib) use weighting function
    //TODO(thib) which eigenvectors should be selected ? extreme of maximal ?
 
public:
    typedef Eigen::SelfAdjointEigenSolver<MatrixType> Solver;
 
protected:
    MatrixType m_cov {MatrixType::Zero()};
    VectorType m_cog {VectorType::Zero()};
    Solver m_solver;    
public:
    PONCA_EXPLICIT_CAST_OPERATORS_DER(NormalCovarianceCurvatureEstimator, normalCovarianceCurvatureEstimator)
    PONCA_FITTING_DECLARE_INIT_ADDDER_FINALIZE
};
 
 
template < class DataPoint, class _NFilter, int DiffType, typename T>
class ProjectedNormalCovarianceCurvatureEstimator : public T
{
    //TODO(thib) check the curvature values that might be wrong
    //TODO(thib) use weighting function
 
PONCA_FITTING_DECLARE_DEFAULT_TYPES
PONCA_FITTING_DECLARE_MATRIX_TYPE
PONCA_FITTING_DECLARE_DEFAULT_DER_TYPES
 
protected:
    enum
    {
        Check = Base::PROVIDES_PRINCIPAL_CURVATURES &&
                Base::PROVIDES_PLANE // \todo This class relies on the primitiveGradient, so update this
    };
 
    enum PASS : int
    {
        FIRST_PASS = 0,
        SECOND_PASS,
        PASS_COUNT
    };
 
public:
    //TODO(thib) use of Eigen::RowAtCompileTime-1 ?
    typedef Eigen::Matrix<Scalar,2,2> Mat22;
    typedef Eigen::Matrix<Scalar,3,2> Mat32;
    typedef Eigen::Matrix<Scalar,2,1> Vector2;
    typedef typename VectorType::Index Index;
    typedef Eigen::SelfAdjointEigenSolver<Mat22> Solver;
 
protected:
    Vector2 m_cog  {Vector2::Zero()}; 
    Mat22 m_cov    {Mat22::Zero()};   
    PASS m_pass    {FIRST_PASS};      
    Mat32 m_tframe {Mat32::Zero()};   
    Solver m_solver;                  
public:
    PONCA_EXPLICIT_CAST_OPERATORS_DER(ProjectedNormalCovarianceCurvatureEstimator, projectedNormalCovarianceCurvature)
    PONCA_FITTING_DECLARE_INIT_ADDDER_FINALIZE
};
 
#include "curvatureEstimation.hpp"
 
} //namespace Ponca