scispace - formally typeset
Search or ask a question
Proceedings ArticleDOI

ArcFace: Additive Angular Margin Loss for Deep Face Recognition

15 Jun 2019-pp 4690-4699
TL;DR: This paper presents arguably the most extensive experimental evaluation against all recent state-of-the-art face recognition methods on ten face recognition benchmarks, and shows that ArcFace consistently outperforms the state of the art and can be easily implemented with negligible computational overhead.
Abstract: One of the main challenges in feature learning using Deep Convolutional Neural Networks (DCNNs) for large-scale face recognition is the design of appropriate loss functions that can enhance the discriminative power. Centre loss penalises the distance between deep features and their corresponding class centres in the Euclidean space to achieve intra-class compactness. SphereFace assumes that the linear transformation matrix in the last fully connected layer can be used as a representation of the class centres in the angular space and therefore penalises the angles between deep features and their corresponding weights in a multiplicative way. Recently, a popular line of research is to incorporate margins in well-established loss functions in order to maximise face class separability. In this paper, we propose an Additive Angular Margin Loss (ArcFace) to obtain highly discriminative features for face recognition. The proposed ArcFace has a clear geometric interpretation due to its exact correspondence to geodesic distance on a hypersphere. We present arguably the most extensive experimental evaluation against all recent state-of-the-art face recognition methods on ten face recognition benchmarks which includes a new large-scale image database with trillions of pairs and a large-scale video dataset. We show that ArcFace consistently outperforms the state of the art and can be easily implemented with negligible computational overhead. To facilitate future research, the code has been made available.

Content maybe subject to copyright    Report

Citations
More filters
Proceedings ArticleDOI
01 Jan 2022
TL;DR: Zhang et al. as discussed by the authors proposed a deep feature prior guided network (DFPGnet) to restore facial details using the estimated deep feature priors from a blurred image.
Abstract: Most recent face deblurring methods have focused on utilizing facial shape priors such as face landmarks and parsing maps. While these priors can provide facial geometric cues effectively, they are insufficient to contain local texture details that act as important clues to solve face deblurring problem. To deal with this, we focus on estimating the deep features of pre-trained face recognition networks (e.g., VGGFace network) that include rich information about sharp faces as a prior, and adopt a generative adversarial network (GAN) to learn it. To this end, we propose a deep feature prior guided network (DFPGnet) that restores facial details using the estimated the deep feature prior from a blurred image. In our DFPGnet, the generator is divided into two streams including prior estimation and deblurring streams. Since the estimated deep features of the prior estimation stream are learned from the VGGFace network which is trained for face recognition not for deblurring, we need to alleviate the discrepancy of feature distributions between the two streams. Therefore, we present feature transform modules at the connecting points of the two streams. In addition, we propose a channel-attention feature discriminator and prior loss, which encourages the generator to focus on more important channels for deblurring among the deep feature prior during training. Experimental results show that our method achieves state-of-the-art performance both qualitatively and quantitatively.

4 citations

Journal ArticleDOI
TL;DR: In this article , a web-based, remote visual discrimination experiment on the basis of principles adopted from the field of psychophysics was conducted with the aim of examining human proficiency in detecting different types of digitally manipulated face images, specifically face swapping, morphing, and retouching.
Abstract: In recent years, increasing deployment of face recognition technology in security-critical settings, such as border control or law enforcement, has led to considerable interest in the vulnerability of face recognition systems to attacks utilising legitimate documents, which are issued on the basis of digitally manipulated face images. As automated manipulation and attack detection remains a challenging task, conventional processes with human inspectors performing identity verification remain indispensable. These circumstances merit a closer investigation of human capabilities in detecting manipulated face images, as previous work in this field is sparse and often concentrated only on specific scenarios and biometric characteristics. This work introduces a web-based, remote visual discrimination experiment on the basis of principles adopted from the field of psychophysics and subsequently discusses interdisciplinary opportunities with the aim of examining human proficiency in detecting different types of digitally manipulated face images, specifically face swapping, morphing, and retouching. In addition to analysing appropriate performance measures, a possible metric of detectability is explored. Experimental data of 306 probands indicate that detection performance is widely distributed across the population and detection of certain types of face image manipulations is much more challenging than others.

4 citations

Proceedings ArticleDOI
01 Sep 2019
TL;DR: The proposed unsupervised approach can effectively model both structural and temporal information of face features in target videos using structural and dynamical dictionaries, respectively, and effectively learn robust and discriminative representations for videos and improve the face verification performance.
Abstract: We propose a hybrid dictionary learning and matching approach using deep features for unconstrained videobased face verification. Popular off-the-shelf image-based deep neural networks often fail to effectively exploit multiple frames for video-based verification. Unlike recurrent neural network-based approaches which require an external large-scale annotated data for training, the proposed unsupervised approach can effectively model both structural and temporal information of face features in target videos using structural and dynamical dictionaries, respectively. We propose an iterative optimization procedure to learn the dynamical dictionaries from videos. Using the learned dictionaries, we model video-to-video similarity as subspace-to-subspace similarity which is not only more robust but also utilizes the information in multiple frames better than the widely used reconstruction error-based measures, where the subspaces are spanned by the learned dictionaries. Experiments on challenging video-based face verification datasets, including Multiple Biometric Grand Challenge (MBGC), Face and Ocular Challenge Series (FOCS) and IARPA JANUS Benchmark A (IJB-A) datasets, demonstrate that the proposed method can effectively learn robust and discriminative representations for videos and improve the face verification performance.

4 citations


Cites methods from "ArcFace: Additive Angular Margin Lo..."

  • ...We also apply our method on features from the state-of-the-art network ArcFace [8]; results from ArcFace are shown with the prefix Arc-....

    [...]

  • ...Similar to MBGC, we also apply our method on the ArcFace features, the results of which are shown with the prefix Arc-....

    [...]

  • ...Crystal loss features perform better than ArcFace features in more challenging protocols like activity-vs-walking and activity-vs-activity....

    [...]

  • ...Experiments with ArcFace features are underway and the results will be presented in a subsequent publication....

    [...]

  • ...• The results using crystal loss features are comparable to the results using ArcFace features....

    [...]

Journal ArticleDOI
TL;DR: Wang et al. as mentioned in this paper proposed POSTER++, which improves POSTER in three directions: cross-fusion, two-stream, and multi-scale feature extraction.
Abstract: Facial expression recognition (FER) plays an important role in a variety of real-world applications such as human-computer interaction. POSTER achieves the state-of-the-art (SOTA) performance in FER by effectively combining facial landmark and image features through two-stream pyramid cross-fusion design. However, the architecture of POSTER is undoubtedly complex. It causes expensive computational costs. In order to relieve the computational pressure of POSTER, in this paper, we propose POSTER++. It improves POSTER in three directions: cross-fusion, two-stream, and multi-scale feature extraction. In cross-fusion, we use window-based cross-attention mechanism replacing vanilla cross-attention mechanism. We remove the image-to-landmark branch in the two-stream design. For multi-scale feature extraction, POSTER++ combines images with landmark's multi-scale features to replace POSTER's pyramid design. Extensive experiments on several standard datasets show that our POSTER++ achieves the SOTA FER performance with the minimum computational cost. For example, POSTER++ reached 92.21% on RAF-DB, 67.49% on AffectNet (7 cls) and 63.77% on AffectNet (8 cls), respectively, using only 8.4G floating point operations (FLOPs) and 43.7M parameters (Param). This demonstrates the effectiveness of our improvements.

4 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

Journal Article
TL;DR: It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.
Abstract: Deep neural nets with a large number of parameters are very powerful machine learning systems. However, overfitting is a serious problem in such networks. Large networks are also slow to use, making it difficult to deal with overfitting by combining the predictions of many different large neural nets at test time. Dropout is a technique for addressing this problem. The key idea is to randomly drop units (along with their connections) from the neural network during training. This prevents units from co-adapting too much. During training, dropout samples from an exponential number of different "thinned" networks. At test time, it is easy to approximate the effect of averaging the predictions of all these thinned networks by simply using a single unthinned network that has smaller weights. This significantly reduces overfitting and gives major improvements over other regularization methods. We show that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.

33,597 citations

Proceedings Article
Sergey Ioffe1, Christian Szegedy1
06 Jul 2015
TL;DR: Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin.
Abstract: Training Deep Neural Networks is complicated by the fact that the distribution of each layer's inputs changes during training, as the parameters of the previous layers change. This slows down the training by requiring lower learning rates and careful parameter initialization, and makes it notoriously hard to train models with saturating nonlinearities. We refer to this phenomenon as internal covariate shift, and address the problem by normalizing layer inputs. Our method draws its strength from making normalization a part of the model architecture and performing the normalization for each training mini-batch. Batch Normalization allows us to use much higher learning rates and be less careful about initialization, and in some cases eliminates the need for Dropout. Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin. Using an ensemble of batch-normalized networks, we improve upon the best published result on ImageNet classification: reaching 4.82% top-5 test error, exceeding the accuracy of human raters.

30,843 citations

28 Oct 2017
TL;DR: An automatic differentiation module of PyTorch is described — a library designed to enable rapid research on machine learning models that focuses on differentiation of purely imperative programs, with a focus on extensibility and low overhead.
Abstract: In this article, we describe an automatic differentiation module of PyTorch — a library designed to enable rapid research on machine learning models. It builds upon a few projects, most notably Lua Torch, Chainer, and HIPS Autograd [4], and provides a high performance environment with easy access to automatic differentiation of models executed on different devices (CPU and GPU). To make prototyping easier, PyTorch does not follow the symbolic approach used in many other deep learning frameworks, but focuses on differentiation of purely imperative programs, with a focus on extensibility and low overhead. Note that this preprint is a draft of certain sections from an upcoming paper covering all PyTorch features.

13,268 citations

Posted Content
TL;DR: The TensorFlow interface and an implementation of that interface that is built at Google are described, which has been used for conducting research and for deploying machine learning systems into production across more than a dozen areas of computer science and other fields.
Abstract: TensorFlow is an interface for expressing machine learning algorithms, and an implementation for executing such algorithms. A computation expressed using TensorFlow can be executed with little or no change on a wide variety of heterogeneous systems, ranging from mobile devices such as phones and tablets up to large-scale distributed systems of hundreds of machines and thousands of computational devices such as GPU cards. The system is flexible and can be used to express a wide variety of algorithms, including training and inference algorithms for deep neural network models, and it has been used for conducting research and for deploying machine learning systems into production across more than a dozen areas of computer science and other fields, including speech recognition, computer vision, robotics, information retrieval, natural language processing, geographic information extraction, and computational drug discovery. This paper describes the TensorFlow interface and an implementation of that interface that we have built at Google. The TensorFlow API and a reference implementation were released as an open-source package under the Apache 2.0 license in November, 2015 and are available at www.tensorflow.org.

10,447 citations