Reconstrucción facial 3D mediante cámara de profundidad RGB-D

Facial detection and recognition systems are a type of technology under development and have found many applications in today society. As an example of one of its many applications are inside the smart phones (smartphones) for computer security. Another example is present in the security cameras that are frequently installed in cities to be able to electronic surveillance in the streets with the advance of these technologies have developed cameras that can give us not only an image in two dimensions, but also a measure of depth. In the present project, a prototype of a code is being developed which is focused on facial detection, its reconstruction of images in three dimensions and the application of filters to eliminate the points of noise within the images analyzed and their smoothing. The project has been developed in a digital environment, where the Python environment has been used for its programming. The use of an Orbbec Astra Pro depth camera to obtain an image with the respective depth of each of its parts in order to be able to represent it in 3D. Finally, a graphic interface and a functional prototype were designed to carry out field tests. The experimental results are satisfactory, demonstrating that the proposed algorithms and the RGB-D camera, which provides a large number of significant points, are a feasible tool for the recognition and reconstruction of images.