void FaceDetector::writeFaceDetectionResult(const ed::Measurement& msr, const cv::Rect& rgb_roi,
const std::vector<cv::Rect>& rgb_face_rois,
int& face_counter, tue::Configuration& result) const
{
// get color image
const cv::Mat& color_image = msr.image()->getRGBImage();
// get color image
const cv::Mat& depth_image = msr.image()->getDepthImage();
// Calculate size factor between depth and rgb images
double f_depth_rgb = (double)depth_image.cols / color_image.cols;
// Create depth view
rgbd::View depth_view(*msr.image(), depth_image.cols);
for (uint j = 0; j < rgb_face_rois.size(); j++)
{
cv::Rect rgb_face_roi = rgb_face_rois[j];
rgb_face_roi.x += rgb_roi.x;
rgb_face_roi.y += rgb_roi.y;
if (debug_mode_)
ed::perception::saveDebugImage("face_detector-rgb", color_image(rgb_face_roi));
result.addArrayItem();
result.setValue("index", face_counter);
// add 2D location of the face
result.setValue("x", rgb_face_roi.x);
result.setValue("y", rgb_face_roi.y);
result.setValue("width", rgb_face_roi.width);
result.setValue("height", rgb_face_roi.height);
// Compute face roi for depth image
cv::Rect depth_face_roi(f_depth_rgb * rgb_face_roi.x, f_depth_rgb * rgb_face_roi.y,
f_depth_rgb * rgb_face_roi.width, f_depth_rgb * rgb_face_roi.height);
if (debug_mode_)
ed::perception::saveDebugImage("face_detector-depth", depth_image(depth_face_roi));
cv::Mat face_area = depth_image(depth_face_roi);
float avg_depth = ed::perception::getMedianDepth(face_area);
if (avg_depth > 0)
{
// calculate the center point of the face
cv::Point2i p_2d(depth_face_roi.x + depth_face_roi.width/2,
depth_face_roi.y + depth_face_roi.height/2);
geo::Vector3 projection = depth_view.getRasterizer().project2Dto3D(p_2d.x, p_2d.y) * avg_depth;
geo::Vector3 point_map = msr.sensorPose() * projection;
// add 3D location of the face
result.setValue("map_x", point_map.x);
result.setValue("map_y", point_map.y);
result.setValue("map_z", point_map.z);
}
else
{
std::cout << "[ED FACE DETECTOR] Could not calculate face's average depth. Map coordinates might be incorrect!" << std::endl;
}
result.endArrayItem();
face_counter++;
}
}
void HumanContourMatcher::process(ed::EntityConstPtr e, tue::Configuration& result) const
{
// if initialization failed, return
if (!init_success_)
return;
// Get the best measurement from the entity
ed::MeasurementConstPtr msr = e->lastMeasurement();
if (!msr)
return;
float depth_sum = 0;
float avg_depht;
float classification_error = 0;
float classification_deviation = 0;
std::string classification_stance;
uint point_counter = 0;
bool is_human = false;
// Get the depth image from the measurement
const cv::Mat& depth_image = msr->image()->getDepthImage();
const cv::Mat& color_image = msr->image()->getRGBImage();
cv::Mat mask_cv = cv::Mat::zeros(depth_image.rows, depth_image.cols, CV_8UC1);
// Iterate over all points in the mask. You must specify the width of the image on which you
// want to apply the mask (see the begin(...) method).
for(ed::ImageMask::const_iterator it = msr->imageMask().begin(depth_image.cols); it != msr->imageMask().end(); ++it)
{
// mask's (x, y) coordinate in the depth image
const cv::Point2i& p_2d = *it;
// TODO dont creat a Mat mask, just give human_classifier_.Classify a vector of 2D points!
// paint a mask
mask_cv.at<unsigned char>(p_2d) = 255;
// calculate measurement average depth
depth_sum += depth_image.at<float>(p_2d);
point_counter++;
}
avg_depht = depth_sum/(float)point_counter;
is_human = human_classifier_.Classify(depth_image, color_image, mask_cv, avg_depht, classification_error, classification_deviation, classification_stance);
// ----------------------- assert results -----------------------
// create group if it doesnt exist
if (!result.readGroup("perception_result", tue::OPTIONAL))
{
result.writeGroup("perception_result");
}
// assert the results to the entity
result.writeGroup(kModuleName);
result.setValue("label", "Human Shape");
if (classification_error > 0){
result.setValue("stance", classification_stance);
result.setValue("error", classification_error);
result.setValue("deviation", classification_deviation);
}
if(is_human)
{
result.setValue("score", 1.0);
}else{
result.setValue("score", 0.0);
}
result.endGroup(); // close human_contour_matcher group
result.endGroup(); // close perception_result group
}