Deep Face Recognition: A Survey

Abstract

Deep learning applies multiple processing layers to learn representations of data with multiple levels of feature extraction. This emerging technique has reshaped the research landscape of face recognition (FR) since 2014, launched by the breakthroughs of DeepFace and DeepID. Since then, deep learning technique, characterized by the hierarchical architecture to stitch together pixels into invariant face representation, has dramatically improved the state-of-the-art performance and fostered successful real-world applications. In this survey, we provide a comprehensive review of the recent developments on deep FR, covering broad topics on algorithm designs, databases, protocols, and application scenes. First, we summarize different network architectures and loss functions proposed in the rapid evolution of the deep FR methods. Second, the related face processing methods are categorized into two classes: "one-to-many augmentation" and "many-to-one normalization". Then, we summarize and compare the commonly used databases for both model training and evaluation. Third, we review miscellaneous scenes in deep FR, such as cross-factor, heterogenous, multiple-media and industrial scenes. Finally, the technical challenges and several promising directions are highlighted.

arXiv commentNeurocomputing

References in corpus (58)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• FaceNet: A Unified Embedding for Face Recognition and Clustering
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
• Neural Architecture Search with Reinforcement Learning
• Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding
• Deep Visual Domain Adaptation: A Survey
• Squeeze-and-Excitation Networks
• Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1
• BinaryConnect: Training Deep Neural Networks with binary weights during propagations
• Deep Learning Face Representation by Joint Identification-Verification
• Learning Face Representation from Scratch
• PCANet: A Simple Deep Learning Baseline for Image Classification?
• Additive Margin Softmax for Face Verification
• Binarized Neural Networks
• DeepID3: Face Recognition with Very Deep Neural Networks
• ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
• Learning to Discover Cross-Domain Relations with Generative Adversarial Networks
• NormFace: L2 Hypersphere Embedding for Face Verification
• Robust Face Recognition via Multimodal Deep Face Representation
• Trunk-Branch Ensemble Convolutional Neural Networks for Video-based Face Recognition
• Learn Convolutional Neural Network for Face Anti-Spoofing
• Searching for Efficient Multi-Scale Architectures for Dense Image Prediction
• Cross-Age LFW: A Database for Studying Cross-Age Face Recognition in Unconstrained Environments
• L2-constrained Softmax Loss for Discriminative Face Verification
• Weakly-supervised Disentangling with Recurrent Transformations for 3D View Synthesis
• Multi-Task Convolutional Neural Network for Pose-Invariant Face Recognition
• Targeting Ultimate Accuracy: Face Recognition via Deep Embedding
• On the Reconstruction of Face Images from Deep Face Templates
• Naive-Deep Face Recognition: Touching the Limit of LFW Benchmark or Not?
• Towards Transferable Adversarial Attack against Deep Face Recognition
• Triplet Probabilistic Embedding for Face Verification and Clustering
• Cross-Domain Visual Matching via Generalized Similarity Measure and Feature Learning
• End-to-End Photo-Sketch Generation via Fully Convolutional Representation Learning
• Rethinking Feature Discrimination and Polymerization for Large-scale Recognition
• Recover Canonical-View Faces in the Wild with Deep Neural Networks
• One-shot Face Recognition by Promoting Underrepresented Classes
• Face Search at Scale: 80 Million Gallery
• Learning from Millions of 3D Scans for Large-scale 3D Face Recognition
• Towards End-to-End Face Recognition through Alignment Learning
• Unravelling Robustness of Deep Learning based Face Recognition Against Adversarial Attacks
• von Mises-Fisher Mixture Model-based Deep learning: Application to Face Verification
• Range Loss for Deep Face Recognition with Long-tail
• WebCaricature: a benchmark for caricature recognition
• Content-Adaptive Sketch Portrait Generation by Decompositional Representation Learning
• Triplet Similarity Embedding for Face Verification
• Towards Large-Pose Face Frontalization in the Wild
• Face Recognition via Centralized Coordinate Learning
• A Good Practice Towards Top Performance of Face Recognition: Transferred Deep Feature Fusion
• Deeply Coupled Auto-encoder Networks for Cross-view Classification
• Feature Transfer Learning for Deep Face Recognition with Under-Represented Data
• Learning a Metric Embedding for Face Recognition using the Multibatch Method
• How to Train Triplet Networks with 100K Identities?
• On the Capacity of Face Representation
• SensitiveLoss: Improving Accuracy and Fairness of Face Representations with Discrimination-Aware Deep Learning
• Pairwise Relational Networks for Face Recognition
• GridFace: Face Rectification via Learning Local Homography Transformations
• CNN-based Real-time Dense Face Reconstruction with Inverse-rendered Photo-realistic Face Images

Cited by in corpus (66)

• A Survey of Deep Learning-based Object Detection
• Medical Deep Learning -- A systematic Meta-Review
• Machine Learning for Microcontroller-Class Hardware: A Review
• How Deep Learning Sees the World: A Survey on Adversarial Attacks & Defenses
• Cross-Resolution Learning for Face Recognition
• Deep Learning Based Single Sample Per Person Face Recognition: A Survey
• Meta Balanced Network for Fair Face Recognition
• Fast Facial Landmark Detection and Applications: A Survey
• Crystal structure prediction with machine learning-based element substitution
• Disentangling Identity and Pose for Facial Expression Recognition
• Detection of Face Recognition Adversarial Attacks
• Relational Deep Feature Learning for Heterogeneous Face Recognition
• The OARF Benchmark Suite: Characterization and Implications for Federated Learning Systems
• LFFD: A Light and Fast Face Detector for Edge Devices
• About Face: A Survey of Facial Recognition Evaluation
• OPOM: Customized Invisible Cloak towards Face Privacy Protection
• Mitigate Bias in Face Recognition using Skewness-Aware Reinforcement Learning
• Racial Faces in-the-Wild: Reducing Racial Bias by Information Maximization Adaptation Network
• Sejong Face Database: A Multi-Modal Disguise Face Database
• GaitFormer: Learning Gait Representations with Noisy Multi-Task Learning
• Unveiling the frontiers of deep learning: innovations shaping diverse domains
• Second Edition FRCSyn Challenge at CVPR 2024: Face Recognition Challenge in the Era of Synthetic Data
• Adjusting Logit in Gaussian Form for Long-Tailed Visual Recognition
• Second FRCSyn-onGoing: Winning Solutions and Post-Challenge Analysis to Improve Face Recognition with Synthetic Data
• This Person (Probably) Exists. Identity Membership Attacks Against GAN Generated Faces
• MaskFace: multi-task face and landmark detector
• Cancelable Biometric Template Generation Using Random Feature Vector Transformations
• Impact of Facial Tattoos and Paintings on Face Recognition Systems
• Fine-Grained Image Analysis with Deep Learning: A Survey
• FairIF: Boosting Fairness in Deep Learning via Influence Functions with Validation Set Sensitive Attributes
• Minimalist Vision with Freeform Pixels
• FaceHack: Triggering backdoored facial recognition systems using facial characteristics
• Prototype Memory for Large-scale Face Representation Learning
• Comparing Human and Machine Bias in Face Recognition
• Unified Face Matching and Physical-Digital Spoofing Attack Detection
• Adversarial Learning with Margin-based Triplet Embedding Regularization
• PropagationNet: Propagate Points to Curve to Learn Structure Information
• Darker than Black-Box: Face Reconstruction from Similarity Queries
• Inclusive Portraits: Race-Aware Human-in-the-Loop Technology
• GB-CosFace: Rethinking Softmax-based Face Recognition from the Perspective of Open Set Classification
• Data-Driven Forecasting of Non-Equilibrium Solid-State Dynamics
• Don't Generate Me: Training Differentially Private Generative Models with Sinkhorn Divergence
• M^2 Deep-ID: A Novel Model for Multi-View Face Identification Using Convolutional Deep Neural Networks
• From Pixels to Words: Leveraging Explainability in Face Recognition through Interactive Natural Language Processing
• Large Scale Open-Set Deep Logo Detection
• Inter-class Discrepancy Alignment for Face Recognition
• Selective Pseudo-Labeling with Reinforcement Learning for Semi-Supervised Domain Adaptation
• Representation based and Attention augmented Meta learning
• ComFace: Facial Representation Learning with Synthetic Data for Comparing Faces
• Generating 2D and 3D Master Faces for Dictionary Attacks with a Network-Assisted Latent Space Evolution
• Lagrangian Objective Function Leads to Improved Unforeseen Attack Generalization in Adversarial Training
• Determination of the most representative descriptor among a set of feature vectors for the same object
• Bandwidth Slicing to Boost Federated Learning in Edge Computing
• Coupling deep and handcrafted features to assess smile genuineness
• Multi Modal Adaptive Normalization for Audio to Video Generation
• Robust One Shot Audio to Video Generation
• One Shot Audio to Animated Video Generation
• Evaluation of Human and Machine Face Detection using a Novel Distinctive Human Appearance Dataset
• Bayesian Metric Learning for Uncertainty Quantification in Image Retrieval
• Human Recognition Using Face in Computed Tomography
• Balanced Masked and Standard Face Recognition
• Profile to Frontal Face Recognition in the Wild Using Coupled Conditional GAN
• Canonical Saliency Maps: Decoding Deep Face Models
• Application of Facial Recognition using Convolutional Neural Networks for Entry Access Control
• FusiformNet: Extracting Discriminative Facial Features on Different Levels
• Information-Theoretic Bias Assessment Of Learned Representations Of Pretrained Face Recognition