Deep learning and face recognition: the state of the art

Abstract

Deep Neural Networks (DNNs) have established themselves as a dominant technique in machine learning. DNNs have been top performers on a wide variety of tasks including image classification, speech recognition, and face recognition. 1-3 Convolutional neural networks (CNNs) have been used in nearly all of the top performing methods on the Labeled Faces in the Wild (LFW) dataset. [3][4][5][6] In this talk and accompanying paper, I attempt to provide a review and summary of the deep learning techniques used in the state-of-the-art. In addition, I highlight the need for both larger and more challenging public datasets to benchmark these systems.Despite the ability of DNNs and autoencoders to perform unsupervised feature learning, modern facial recognition pipelines still require domain specific engineering in the form of re-alignment. For example, in Facebook's recent DeepFace paper, a 3D "frontalization" step lies at the beginning of the pipeline. This step creates a 3D face model for the incoming image and then uses a series of affine transformations of the fiducial points to "frontalize" the image. This step enables the DeepFace system to use a neural network architecture with locally connected layers without weight sharing as opposed to standard convolutional layers. 6 Deep learning techniques combined with large datasets have allowed research groups to surpass human level performance on the LFW dataset. 3, 5The high accuracy (99.63% for FaceNet at the time of publishing) and utilization of outside data (hundreds of millions of images in the case of Google's FaceNet) suggest that current face verification benchmarks such as LFW may not be challenging enough, nor provide enough data, for current techniques. 3, 5 There exist a variety of organizations with mobile photo sharing applications that would be capable of releasing a very large scale and highly diverse dataset of facial images captured on mobile devices. Such an "ImageNet for Face Recognition" would likely receive a warm welcome from researchers and practitioners alike.

Implementing CCTV-Based Attendance Taking Support System Using Deep Face Recognition: A Case Study at FPT Polytechnic College

“…Recently, deep learning techniques have made many significant achievements in FR, such as deep convolutional neural networks [4] use a cascade of multiple layers of processing units for feature extraction. They learn various levels of representations that correspond to different levels of abstraction.…”

Face Recognition on Low Resolution Face Image With TBE-CNN Architecture

“…Stochastic Gradient Descent (SGD) is an extension of gradient descent where it's a modification of "batch" gradient descent and parameter updates are made after calculating a stochastic approximation of the gradient [20]. Gradient Descent calculate the gradient of cost function as much as number of training data.…”

Feature redundancy approach to efficient face recognition in still images

“…They used PCA and SVM, compared the results with random (uncontrolled and controlled) training samples selection, and evaluated the recognition accuracy of each method. Deep neural network DNNs are fast becoming the dominant technique in machine learning [6]. Their performance has been most successful for a wide variety of tasks, including image classification, speech synthesis, speech recognition, and face recognition.…”

“…Their performance has been most successful for a wide variety of tasks, including image classification, speech synthesis, speech recognition, and face recognition. For a review and application of state-of-the-art deep learning approaches to face recognition, see [6,60,93]…”