Efficient Face Recognition System for Operating in Unconstrained Environments

Abstract

Facial recognition is fundamental for a wide variety of security systems operating in real-time applications. Recently, several deep neural networks algorithms have been developed to achieve state-of-the-art performance on this task. The present work was conceived due to the need for an efficient and low-cost processing system, so a real-time facial recognition system was proposed using a combination of deep learning algorithms like FaceNet and some traditional classifiers like SVM, KNN, and RF using moderate hardware to operate in an unconstrained environment. Generally, a facial recognition system involves two main tasks: face detection and recognition. The proposed scheme uses the YOLO-Face method for the face detection task which is a high-speed real-time detector based on YOLOv3, while, for the recognition stage, a combination of FaceNet with a supervised learning algorithm, such as the support vector machine (SVM), is proposed for classification. Extensive experiments on unconstrained datasets demonstrate that YOLO-Face provides better performance when the face under an analysis presents partial occlusion and pose variations; besides that, it can detect small faces. The face detector was able to achieve an accuracy of over 89.6% using the Honda/UCSD dataset which runs at 26 FPS with darknet-53 to VGA-resolution images for classification tasks. The experimental results have demonstrated that the FaceNet+SVM model was able to achieve an accuracy of 99.7% using the LFW dataset. On the same dataset, FaceNet+KNN and FaceNet+RF achieve 99.5% and 85.1%, respectively; on the other hand, the FaceNet was able to achieve 99.6%. Finally, the proposed system provides a recognition accuracy of 99.1% and 49 ms runtime when both the face detection and classifications stages operate together.

Tracking Missing Person in Large Crowd Gathering Using Intelligent Video Surveillance

“…According to Sanchez-Moreno, A.S. et al. [ 7 ], some deep neural networks techniques have recently been created to attain state-of-the-art performance on tracking of missing person through face detection and recognition problem. Their work is not for a densely populated environment.…”

“…As we noticed in Table 1 , the authors in [ 7 ] applied state-of-art deep learning techniques for face detection and recognition using conversion of low resolution images to high-quality images, but the technique is not tested in low-resolution images from large gatherings. Moreover, literature in [ 1 , 2 , 3 , 4 , 5 , 11 ] shows work on recognizing people based on large crowd and low resolution image data, whereas the literature presented in [ 12 ] only depicts exploitation of large crowd data and in [ 13 ] research carried out only on low resolution data.…”

Comparative analysis of classifiers for face recognition on image fragments identified by the FaceNet neural network

“…The paper devoted to neural network methods for face recognition on image fragments [5,13]. A specificity of these methods is the automatic extraction of image features and the mapping of inputs to network outputs as a result of its training.…”

“…So, after training on the Labeled Faces in the Wild dataset [15], which contains approximately 3 million images, the percentage of correct recognition for the FaceNet network on the same dataset was 99.42 %, and on the YouTube Faces Database it was 95.12 % [9,14]. In [5], to improve the face recognition performance it was proposed to apply a k-nearest neighbor classifier, a linear support vector machine (SVM), and a random forest to image feature vectors obtained using FaceNet. Compared to the thresholding, higher recognition performance was obtained on the Labeled Faces in the Wild image database [15]; especially for the first two classifiers [5].…”

“…The first one is the face detection on the image. It is usually solved by scanning video stream frames with a window smaller than the frame size [4,5]. In this case, computational difficulties arise due to the using of windows of different locations and sizes [6,7].…”