Efficient Real-Time Face Recognition with Adaptive Boosting Artificial Neural Networks

One of the most important issues at current scenario is the recognition different faces in images. Correct facial recognition will help in areas like person verification and national defense. This project demonstrates the creation of a system that uses the AdaBoost algorithm and artificial neural network technique to recognize a person from their face. The training and testing phases of the system's implementation are divided into two parts. A series of photos is saved during the training phase, and face embeddings are generated for each image and stored as a matrix. Face embeddings from the live webcam image are generated during the testing phase and contrasted with the training image matrix. An artificial neural network is used for this. Compared to other methods, this method is the efficient one. The efficiency is in between 90% and 92%. The software developed will work with multiple faces.