Implements a wrapper around the pre-trained caffe model by Krafka2016. More...

Public Member Functions
	ITracker (const std::string &model_file, const std::string &weights_file, const std::string &mean_left_eye_file, const std::string &mean_right_eye_file, const std::string &mean_face_file)

cv::Vec2f	predict (const cv::Mat &left_eye, const cv::Mat &right_eye, const cv::Mat &face, const cv::Mat &face_mask)

Detailed Description

Implements a wrapper around the pre-trained caffe model by Krafka2016.

Definition at line 31 of file itracker.h.

Constructor & Destructor Documentation

Initializes the iTracker.

Parameters

model_file	A path to the model file, usually a *_deploy.prototxt
weights_file	A path to the trained weights, usually a *.caffemodel
mean_left_file	Path to a *.binaryproto file, holds the mean of left eyes of the training set
mean_right_file	Path to a *.binaryproto file, holds the mean of right eyes of the training set
mean_face_file	Path to a *.binaryprotofile, holds the mean of faces of the training set

Definition at line 140 of file itracker.cpp.

   : impl(new ITrackerImpl(model_file, weights_file, mean_left_eye_file,
                           mean_left_eye_file, mean_face_file)) {
 }

cv::Vec2f itracker::ITracker::predict	(	const cv::Mat &	left_eye,
		const cv::Mat &	right_eye,
		const cv::Mat &	face,
		const cv::Mat &	face_mask
	)

Performs a forward step for the neural network.

Parameters

left_eye	A cropped left eye.
right_eye	A cropped right eye.
face	The cropped face
face_mask	A 25x25 matrix which has 1 where the face was in the original image (relative), and 0 at the other places

Definition at line 149 of file itracker.cpp.

                                                                        {
   return this->impl->predict(left_eye, right_eye, face, face_mask);
 }

The documentation for this class was generated from the following files: