The theta_face_recognition package uses Freenect_stack and Openni_camera to start the Kinect and be able to use its image topics. We use Dlib to perform face recognition and comparison in the images.
-
kinect_image_node:: This node uses the topic provided by the Kinect and converting between ROS images and OpenCV images, thus functioning as a bridge. -
face_detection_node:: This node uses the topic provided by the kinect_image_node and saves a frame into a .jpg image, being a picture of the Operator. -
face_recognition_node.:: This node also uses the topic provided by the kinect_image_node and saves a frame into a image, with the photo to be compared as Comparador.png. -
trainer_node.:: This node compares the image of the Operator with the Comparador and returns whether the Operator was found in the other image.