Deep learning-based face detection and recognition system

Abstract EN

Face detection and recognition systems have recently achieved encouraging results using deep learning especially Convolutional Neural Network (CNN).

Face detection and recognition system have many challenges in unconstrained environments that decrease the accuracy, for overcoming these challenges a deep learning-based features combination has been proposed for face recognition.

The scheme performs feature-level combination by applying two pre-trained InceptionNet-v1 and VggNet-16 models as deep feature extractors.

First, faces are detected and aligned using Multi-Task Convolutional Neural Networks (MTCNN) face detector then the deep features are extracted from a face image using each individually pre-trained CNN.

Second, features obtained from InceptionNet-v1 and VggNet-16 models are combined using the serial-feature combinations method.

Finally, a classification task is perform using a multiclass Support Vector Machine (SVM) classifier.

Experiments on the following datasets: VggFace2, LFW, Essex, and ORL, indicate the efficacy of the proposed system as the combination of the two pretrained CNN models improves performance.

The combination strategy, in particular, yields an accuracy of 95.33% to 99.29% on all datasets.