Face Recognition using OpenCV, DeepLearning4J with Java - Code Included

TechInDeep: https://www.techindeep.com/

New: Learn how to build neural networks from scratch tutorial series:
Part 1: https://code-ai.mk/neural-networks-making-predictions/
Part 2: https://code-ai.mk/neural-network-making-predictions-part-2/

Face Recognition project done using Java, DeepLearning4J and OpenCV.

Tutorial with Source Code: http://code-ai.mk/?p=29

A facial recognition system is a technology capable of identifying or verifying a person from a digital image or a video frame from a video source. There are multiple methods in which facial recognition systems work, but in general, they work by comparing selected facial features from given image with faces within a database.

In this case an image is selected. Using OpenCV the faces are extracted. When the "Train" button is clicked a pre-trained CNN extracts the features of that face and saves them to a list in memory. A pre-trained network weights from "VGGFACE" are loaded and used to extract the face features from the "pool4" layer of the network. When the "Recognize" button is clicked a distance metric such as the Euclidean distance is used to measure the distance between face points. For each analyzed image the closest point is chosen as a winner. Using this type of architecture allows us to train on any number of faces.

This project is done using asynchronous tasks in JavaFX.

Tutorial and project files are available upon request. This would be the first project in Face Detection/Recognition series using Java