TrainNodeCvGMM.h

// Gaussian Mixture Model (based on OpenCV) training class interface
// Written by Sergey G. Kosov in 2012 for Project X
#pragma once

#include "TrainNode.h"

namespace DirectGraphicalModels
{
    class CSamplesAccumulator;
    
    typedef struct TrainNodeCvGMMParams {
        word    numGausses;                 
        int     covariance_matrix_type;     
        int     maxCount;                   
        double  epsilon;                    
        int     term_criteria_type;         
        size_t  maxSamples;                 

        TrainNodeCvGMMParams() {}
        TrainNodeCvGMMParams(word _numGausses, int _covariance_matrix_type, int _maxCount, double _epsilon, int _term_criteria_type, int _maxSamples) : numGausses(_numGausses), covariance_matrix_type(_covariance_matrix_type), maxCount(_maxCount), epsilon(_epsilon), term_criteria_type(_term_criteria_type), maxSamples(_maxSamples) {}
    } TrainNodeCvGMMParams;

    const TrainNodeCvGMMParams TRAIN_NODE_CV_GMM_PARAMS_DEFAULT = TrainNodeCvGMMParams(
                                                                16,                                         // Number of Gaussians
                                                                ml::EM::COV_MAT_DIAGONAL,                   // Covariance matrix type
                                                                100,                                        // Max number of iterations
                                                                0.01,                                       // GMM accuracy
                                                                TermCriteria::MAX_ITER | TermCriteria::EPS, // Termination cirteria (according the the two previous parameters)
                                                                0                                           // Maximum number of samples to be used in training. 0 means using all the samples                                                      
                                                                );

    // =========================== OpenCV GMM Train Class ===========================
    class CTrainNodeCvGMM : public CTrainNode
    {
    public:
        DllExport CTrainNodeCvGMM(byte nStates, word nFeatures, TrainNodeCvGMMParams params = TRAIN_NODE_CV_GMM_PARAMS_DEFAULT);
        DllExport CTrainNodeCvGMM(byte nStates, word nFeatures, size_t maxSamples, byte nGausses = TRAIN_NODE_CV_GMM_PARAMS_DEFAULT.numGausses);
        DllExport virtual ~CTrainNodeCvGMM(void);

        DllExport void  reset(void);    
        DllExport void  save(const std::string &path, const std::string &name = std::string(), short idx = -1) const;
        DllExport void  load(const std::string &path, const std::string &name = std::string(), short idx = -1);

        DllExport void  addFeatureVec(const Mat &featureVector, byte gt);   
        
        DllExport void  train(bool doClean = false);


    protected:
        DllExport void  saveFile(FILE *pFile) const { } 
        DllExport void  loadFile(FILE *pFile) { } 
        DllExport void  calculateNodePotentials(const Mat &featureVector, Mat &potential, Mat &mask) const;


    private:
        void          init(TrainNodeCvGMMParams params);        // This function is called by both constructors

    
    private:
        static const double MIN_COEFFICIENT_BASE;

    
    protected:
        std::vector<Ptr<ml::EM>>      m_vpEM;                   
        CSamplesAccumulator         * m_pSamplesAcc;            

    private:
        long double                   m_minCoefficient;         // = 1;         // auxilary coefficient for scaling gaussian coefficients
    };
}