RnD-Pedestrian-Detection

Cascadded Classifier

This is an open implimentation of Cascaded Classifier using [1].

Functions:

  1. BoostedCascade(CvStatModel classifier) Done;
  2. ~BoostedCascade() Done;
  3. void clear();
  4. void save( const char* filename, const char* name=0 );
  5. void load( const char* filename, const char* name=0 );
  6. void write( CvFileStorage* storage, const char* name );
  7. void read( CvFileStorage* storage, CvFileNode* node );
  8. bool train(const Mat& train_data, const Mat& responses, int tflag = CV_COL_SAMPLE, const Mat& var_idx =0, const Mat& sample_idx = 0) Done;
  9. float predict(const Mat& sample) const Done;

#References [1] Lienhart, Rainer, Alexander Kuranov, and Vadim Pisarevsky. "Empirical analysis of detection cascades of boosted classifiers for rapid object detection." In Pattern Recognition, pp. 297-304. Springer Berlin Heidelberg, 2003.