scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Deep face recognition: A survey

14 Mar 2021-Neurocomputing (Elsevier)-Vol. 429, pp 215-244
TL;DR: A comprehensive review of the recent developments on deep face recognition can be found in this paper, covering broad topics on algorithm designs, databases, protocols, and application scenes, as well as the technical challenges and several promising directions.
About: This article is published in Neurocomputing.The article was published on 2021-03-14 and is currently open access. It has received 353 citations till now. The article focuses on the topics: Deep learning & Feature extraction.
Citations
More filters
Journal ArticleDOI
TL;DR: In this article, a dual-stream mutual information distillation network (MIND-Net) is proposed to extract non-identity specific mutual information (MI) characterized by generic face features coexisting on realistic and synthetic LR face images to render a resolution-invariant embedding space for LRFR.
Abstract: Realistic low-resolution (LR) face images refer to those captured by the real-world surveillance cameras at extreme standoff distances, thereby LR and poor in quality essentially. Owing to severe scarcity of labeled data, a high-capacity deep convolution neural networks (CNN) is hardly trained to confront the realistic LR face recognition (LRFR) challenge. We introduce in this letter a dual-stream mutual information distillation network (MIND-Net), whereby the non-identity specific mutual information (MI) characterized by generic face features coexistent on realistic and synthetic LR face images are distilled to render a resolution-invariant embedding space for LRFR. For a thorough analysis, we quantify the degree of MI distillation in terms normalized MI index. Our experimental results on the realistic LR face datasets substantiate that the MIND-Net instances assembled from the pre-learned CNNs stand out from the baselines and other state of the arts by a notable margin.

19 citations

Proceedings ArticleDOI
19 Sep 2020
TL;DR: This study discusses the appropriate method to be applied in a presence system using faces by comparing two deep learning architectural models, they are FaceNet and Openface.
Abstract: This study discusses the appropriate method to be applied in a presence system using faces by comparing two deep learning architectural models, they are FaceNet and Openface. FaceNet is a model developed by Google researchers that has the highest accuracy in face recognition. While Openface is a development from FaceNet that is trained with smaller datasets but has an accuracy that is almost equal to FaceNet. This will start by taking the employee's face into an image dataset. From the dataset, the face preprocessing will be performed by detecting, cropping, and resizing the face. Then extracting facial features into 128 dimensions using the FaceNet and Openface. With the Support Vector Machine (SVM), the classification of facial features will be carried out to obtain accuracy. To validate the model, 5 fold cross-validations are used. FaceNet accuracy results that obtained are higher with perfect accuracy that is 100%, while Openface only 93.33% accuracy. The implementation using the model with the highest accuracy (FaceNet) has the same results as the model testing that is 100% using the introduction threshold probability of 0.25.

19 citations

Journal ArticleDOI
TL;DR: Zhang et al. as discussed by the authors proposed new multi-objective genetic programming (GP) algorithms for feature learning in face recognition to achieve effective face feature learning, a new individual representation is developed to allow GP to select informative regions from the input image, extract features using various descriptors, and combine the extracted features for classification.

19 citations

Journal ArticleDOI
TL;DR: The experimental results established that the GWOECN-FR technology outperformed more contemporary approaches in real-time face recognition, and was primarily concerned with reliably and rapidly recognizing faces in input photos.
Abstract: Face recognition (FR) is a technique for recognizing individuals through the use of face photographs. The FR technology is widely applicable in a variety of fields, including security, biometrics, authentication, law enforcement, smart cards, and surveillance. Recent advances in deep learning (DL) models, particularly convolutional neural networks (CNNs), have demonstrated promising results in the field of FR. CNN models that have been pretrained can be utilized to extract characteristics for effective FR. In this regard, this research introduces the GWOECN-FR approach, a unique grey wolf optimization with an enhanced capsule network-based deep transfer learning model for real-time face recognition. The proposed GWOECN-FR approach is primarily concerned with reliably and rapidly recognizing faces in input photos. Additionally, the GWOECN-FR approach is preprocessed in two steps, namely, data augmentation and noise reduction by bilateral filtering (BF). Additionally, for feature vector extraction, an expanded capsule network (ECN) model can be used. Additionally, grey wolf optimization (GWO) combined with a stacked autoencoder (SAE) model is used to identify and classify faces in images. The GWO algorithm is used to optimize the SAE model’s weight and bias settings. The GWOECN-FR technique’s performance is validated using a benchmark dataset, and the results are analyzed in a variety of aspects. The GWOECN-FR approach achieved a TST of 0.03 s on the FEI dataset, whereas the AlexNet-SVM, ResNet-SVM, and AlexNet models achieved TSTs of 0.125 s, 0.0051 s, and 0.0062 s, respectively. The experimental results established that the GWOECN-FR technology outperformed more contemporary approaches.

17 citations

Proceedings ArticleDOI
01 Jun 2022
TL;DR: Wang et al. as discussed by the authors proposed a novel Shuffled Style Assembly Network (SSAN) to extract and reassemble different content and style features for a stylized feature space.
Abstract: With diverse presentation attacks emerging continually, generalizable face anti-spoofing (FAS) has drawn growing attention. Most existing methods implement domain generalization (DG) on the complete representations. However, different image statistics may have unique properties for the FAS tasks. In this work, we separate the complete representation into content and style ones. A novel Shuffled Style Assembly Network (SSAN) is proposed to extract and reassemble different content and style features for a stylized feature space. Then, to obtain a generalized representation, a contrastive learning strategy is developed to emphasize liveness-related style information while suppress the domain-specific one. Finally, the representations of the correct assemblies are used to distinguish between living and spoofing during the inferring. On the other hand, despite the decent performance, there still exists a gap between academia and industry, due to the difference in data quantity and distribution. Thus, a new large-scale benchmark for FAS is built up to further evaluate the performance of algorithms in reality. Both qualitative and quantitative results on existing and proposed benchmarks demonstrate the effectiveness of our methods. The codes will be available at https://github.com/wangzhuo2019/SSAN.

17 citations

References
More filters
Proceedings ArticleDOI
27 Jun 2016
TL;DR: In this article, the authors proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously, which won the 1st place on the ILSVRC 2015 classification task.
Abstract: Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

123,388 citations

Proceedings Article
03 Dec 2012
TL;DR: The state-of-the-art performance of CNNs was achieved by Deep Convolutional Neural Networks (DCNNs) as discussed by the authors, which consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
Abstract: We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0% which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax. To make training faster, we used non-saturating neurons and a very efficient GPU implementation of the convolution operation. To reduce overriding in the fully-connected layers we employed a recently-developed regularization method called "dropout" that proved to be very effective. We also entered a variant of this model in the ILSVRC-2012 competition and achieved a winning top-5 test error rate of 15.3%, compared to 26.2% achieved by the second-best entry.

73,978 citations

Proceedings Article
04 Sep 2014
TL;DR: This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
Abstract: In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

55,235 citations

Proceedings ArticleDOI
07 Jun 2015
TL;DR: Inception as mentioned in this paper is a deep convolutional neural network architecture that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14).
Abstract: We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

40,257 citations

Journal ArticleDOI
08 Dec 2014
TL;DR: A new framework for estimating generative models via an adversarial process, in which two models are simultaneously train: a generative model G that captures the data distribution and a discriminative model D that estimates the probability that a sample came from the training data rather than G.
Abstract: We propose a new framework for estimating generative models via an adversarial process, in which we simultaneously train two models: a generative model G that captures the data distribution, and a discriminative model D that estimates the probability that a sample came from the training data rather than G. The training procedure for G is to maximize the probability of D making a mistake. This framework corresponds to a minimax two-player game. In the space of arbitrary functions G and D, a unique solution exists, with G recovering the training data distribution and D equal to ½ everywhere. In the case where G and D are defined by multilayer perceptrons, the entire system can be trained with backpropagation. There is no need for any Markov chains or unrolled approximate inference networks during either training or generation of samples. Experiments demonstrate the potential of the framework through qualitative and quantitative evaluation of the generated samples.

38,211 citations