Deep Learning Face Representation by Joint Identification-Verification

Abstract

The key challenge of face recognition is to develop effective feature representations for reducing intra-personal variations while enlarging inter-personal differences. In this paper, we show that it can be well solved with deep learning and using both face identification and verification signals as supervision. The Deep IDentification-verification features (DeepID2) are learned with carefully designed deep convolutional networks. The face identification task increases the inter-personal variations by drawing DeepID2 extracted from different identities apart, while the face verification task reduces the intra-personal variations by pulling DeepID2 extracted from the same identity together, both of which are essential to face recognition. The learned DeepID2 features can be well generalized to new identities unseen in the training data. On the challenging LFW dataset, 99.15% face verification accuracy is achieved. Compared with the best deep learning result on LFW, the error rate has been significantly reduced by 67%.

References in corpus (2)

• Large Scale Strongly Supervised Ensemble Metric Learning, with Applications to Face Verification and Retrieval
• Surpassing Human-Level Face Verification Performance on LFW with GaussianFace

Cited by in corpus (198)

• Deep Face Recognition: A Survey
• A Discriminatively Learned CNN Embedding for Person Re-identification
• Robust Face Recognition via Multimodal Deep Face Representation
• Trunk-Branch Ensemble Convolutional Neural Networks for Video-based Face Recognition
• A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects
• Learning Deep Embeddings with Histogram Loss
• Semi-Supervised Sparse Representation Based Classification for Face Recognition with Insufficient Labeled Samples
• Targeting Ultimate Accuracy: Face Recognition via Deep Embedding
• Low-resolution Face Recognition in the Wild via Selective Knowledge Distillation
• Strengths and Weaknesses of Deep Learning Models for Face Recognition Against Image Degradations
• Looking Beyond Appearances: Synthetic Training Data for Deep CNNs in Re-identification
• Deep Adaptive Feature Embedding with Local Sample Distributions for Person Re-identification
• Semantic Image Segmentation via Deep Parsing Network
• On Low-Resolution Face Recognition in the Wild: Comparisons and New Techniques
• Enabling and Emerging Technologies for Social Distancing: A Comprehensive Survey and Open Problems
• Cross-Domain Visual Matching via Generalized Similarity Measure and Feature Learning
• Writer-independent Feature Learning for Offline Signature Verification using Deep Convolutional Neural Networks
• TABOR: A Highly Accurate Approach to Inspecting and Restoring Trojan Backdoors in AI Systems
• SFace: Sigmoid-Constrained Hypersphere Loss for Robust Face Recognition
• Recent Advances in Deep Learning Techniques for Face Recognition
• Personalized Age Progression with Bi-level Aging Dictionary Learning
• Deep learning and face recognition: the state of the art
• Rethinking Softmax Cross-Entropy Loss for Adversarial Robustness
• Efficient Facial Representations for Age, Gender and Identity Recognition in Organizing Photo Albums using Multi-output CNN
• SAR Image Despeckling by Deep Neural Networks: from a pre-trained model to an end-to-end training strategy
• Multi-label Learning Based Deep Transfer Neural Network for Facial Attribute Classification
• Meta Balanced Network for Fair Face Recognition
• Towards End-to-End Face Recognition through Alignment Learning
• Study and Observation of the Variations of Accuracies for Handwritten Digits Recognition with Various Hidden Layers and Epochs using Convolutional Neural Network
• Multi-Representational Learning for Offline Signature Verification using Multi-Loss Snapshot Ensemble of CNNs
• Deep Feature Space: A Geometrical Perspective
• Look, Listen and Learn - A Multimodal LSTM for Speaker Identification
• Minimum Margin Loss for Deep Face Recognition
• Robust Screening of COVID-19 from Chest X-ray via Discriminative Cost-Sensitive Learning
• A Multi-view CNN-based Acoustic Classification System for Automatic Animal Species Identification
• Support Vector Guided Softmax Loss for Face Recognition
• Heatmap Regression via Randomized Rounding
• MiniSeg: An Extremely Minimum Network for Efficient COVID-19 Segmentation
• FingerNet: Pushing The Limits of Fingerprint Recognition Using Convolutional Neural Network
• Mitigate Bias in Face Recognition using Skewness-Aware Reinforcement Learning
• Semi-Latent GAN: Learning to generate and modify facial images from attributes
• Dynamic Curriculum Learning for Imbalanced Data Classification
• When Deep Learning Meets Multi-Task Learning in SAR ATR: Simultaneous Target Recognition and Segmentation
• Racial Faces in-the-Wild: Reducing Racial Bias by Information Maximization Adaptation Network
• Maximum A Posteriori Estimation of Distances Between Deep Features in Still-to-Video Face Recognition
• Striking the Right Balance with Uncertainty
• Implementation of Fruits Recognition Classifier using Convolutional Neural Network Algorithm for Observation of Accuracies for Various Hidden Layers
• Deep Imbalanced Learning for Face Recognition and Attribute Prediction
• Graph-RISE: Graph-Regularized Image Semantic Embedding
• FSCE: Few-Shot Object Detection via Contrastive Proposal Encoding
• Entropic Out-of-Distribution Detection: Seamless Detection of Unknown Examples
• AdaCos: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations
• Improved Sample Complexities for Deep Networks and Robust Classification via an All-Layer Margin
• Self-supervised Multi-view Person Association and Its Applications
• ShrinkTeaNet: Million-scale Lightweight Face Recognition via Shrinking Teacher-Student Networks
• SphereFace2: Binary Classification is All You Need for Deep Face Recognition
• Recent Advances in Deep Learning for Object Detection
• Deep Learning For Face Recognition: A Critical Analysis
• A Performance Comparison of Loss Functions for Deep Face Recognition
• WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognition
• Regularizing Deep Networks with Semantic Data Augmentation
• A Light Dual-Task Neural Network for Haze Removal
• A survey of face recognition techniques under occlusion
• A Deep Four-Stream Siamese Convolutional Neural Network with Joint Verification and Identification Loss for Person Re-detection
• Deep Learning Markov Random Field for Semantic Segmentation
• Transductive Zero-Shot Hashing for Multilabel Image Retrieval
• Cross-Resolution Face Recognition via Prior-Aided Face Hallucination and Residual Knowledge Distillation
• xCos: An Explainable Cosine Metric for Face Verification Task
• Embedding Label Structures for Fine-Grained Feature Representation
• Max-margin Class Imbalanced Learning with Gaussian Affinity
• Deep Metric Learning with Spherical Embedding
• Facial UV Map Completion for Pose-invariant Face Recognition: A Novel Adversarial Approach based on Coupled Attention Residual UNets
• Continual Person Identification using Footstep-Induced Floor Vibrations on Heterogeneous Floor Structures
• Quantum Gradient Algorithm for General Polynomials
• Towards Universal Representation Learning for Deep Face Recognition
• The Impact of Racial Distribution in Training Data on Face Recognition Bias: A Closer Look
• On the Separability of Classes with the Cross-Entropy Loss Function
• Decorrelated Adversarial Learning for Age-Invariant Face Recognition
• NPCFace: Negative-Positive Collaborative Training for Large-scale Face Recognition
• To Balance or Not to Balance: A Simple-yet-Effective Approach for Learning with Long-Tailed Distributions
• Can Synthetic Faces Undo the Damage of Dataset Bias to Face Recognition and Facial Landmark Detection?
• Joint Deep Cross-Domain Transfer Learning for Emotion Recognition
• GroupFace: Learning Latent Groups and Constructing Group-based Representations for Face Recognition
• Masked Face Recognition Challenge: The InsightFace Track Report
• A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
• Video Face Recognition: Component-wise Feature Aggregation Network (C-FAN)
• Towards Unsupervised Domain Adaptation for Deep Face Recognition under Privacy Constraints via Federated Learning
• Character-level Chinese Writer Identification using Path Signature Feature, DropStroke and Deep CNN
• Large-scale Bisample Learning on ID Versus Spot Face Recognition
• Dynamic Region-Aware Convolution
• An Automatic System for Unconstrained Video-Based Face Recognition
• Generative One-Shot Face Recognition
• Heterogeneity Loss to Handle Intersubject and Intrasubject Variability in Cancer
• Zooming into Face Forensics: A Pixel-level Analysis
• Attribute Adaptive Margin Softmax Loss using Privileged Information
• FaceLeaks: Inference Attacks against Transfer Learning Models via Black-box Queries
• Leveraging Angular Distributions for Improved Knowledge Distillation
• Ulixes: Facial Recognition Privacy with Adversarial Machine Learning
• RobFR: Benchmarking Adversarial Robustness on Face Recognition
• BroadFace: Looking at Tens of Thousands of People at Once for Face Recognition
• Hierarchical Feature Embedding for Attribute Recognition
• 3D Face Recognition: A Survey
• Successive Embedding and Classification Loss for Aerial Image Classification
• Learning Meta Face Recognition in Unseen Domains
• Susceptibility to Image Resolution in Face Recognition and Trainings Strategies
• Hybrid-Attention based Decoupled Metric Learning for Zero-Shot Image Retrieval
• Signal-to-Noise Ratio: A Robust Distance Metric for Deep Metric Learning
• Multi-Prototype Networks for Unconstrained Set-based Face Recognition
• Anatomy-Aware Siamese Network: Exploiting Semantic Asymmetry for Accurate Pelvic Fracture Detection in X-ray Images
• Learning Rich Features for Gait Recognition by Integrating Skeletons and Silhouettes
• Adversarial Learning with Margin-based Triplet Embedding Regularization
• Can we still avoid automatic face detection?
• Learning to Cluster Faces via Confidence and Connectivity Estimation
• Towards Backward-Compatible Representation Learning
• Adaptive Selection of Deep Learning Models on Embedded Systems
• Revisiting Deep Neural Network Test Coverage from the Test Effectiveness Perspective
• CL2R: Compatible Lifelong Learning Representations
• GridFace: Face Rectification via Learning Local Homography Transformations
• Text Embeddings for Retrieval From a Large Knowledge Base
• Learning to Learn in a Semi-Supervised Fashion
• A Survey of Deep Facial Attribute Analysis
• Dependency Aware Filter Pruning
• Softmax Dissection: Towards Understanding Intra- and Inter-class Objective for Embedding Learning
• PAM: Pose Attention Module for Pose-Invariant Face Recognition
• DropSample: A New Training Method to Enhance Deep Convolutional Neural Networks for Large-Scale Unconstrained Handwritten Chinese Character Recognition
• Balanced Alignment for Face Recognition: A Joint Learning Approach
• Over-the-Air Adversarial Flickering Attacks against Video Recognition Networks
• Disentanglement for Discriminative Visual Recognition
• Deep Metric Learning by Online Soft Mining and Class-Aware Attention
• Discriminability Distillation in Group Representation Learning
• Asymmetric Rejection Loss for Fairer Face Recognition
• Faceness-Net: Face Detection through Deep Facial Part Responses
• Triplet Distillation for Deep Face Recognition
• Style Transfer Applied to Face Liveness Detection with User-Centered Models
• CoReS: Compatible Representations via Stationarity
• Consensus-Driven Propagation in Massive Unlabeled Data for Face Recognition
• Learning Semantic Vector Representations of Source Code via a Siamese Neural Network
• Learning Channel Inter-dependencies at Multiple Scales on Dense Networks for Face Recognition
• Metric Classification Network in Actual Face Recognition Scene
• Face Synthesis for Eyeglass-Robust Face Recognition
• Improving Face Recognition from Hard Samples via Distribution Distillation Loss
• Searching for Apparel Products from Images in the Wild
• Deep Networks with Fast Retraining
• Learnable Adaptive Cosine Estimator (LACE) for Image Classification
• A Multi-Resolution Word Embedding for Document Retrieval from Large Unstructured Knowledge Bases
• Visual Diver Face Recognition for Underwater Human-Robot Interaction
• An Attention-Based Speaker Naming Method for Online Adaptation in Non-Fixed Scenarios
• On Numerosity of Deep Neural Networks
• A Novel Automatic Modulation Classification Scheme Based on Multi-Scale Networks
• Inter-class Discrepancy Alignment for Face Recognition
• Semi-Siamese Training for Shallow Face Learning
• MassFace: an efficient implementation using triplet loss for face recognition
• DSAM: A Distance Shrinking with Angular Marginalizing Loss for High Performance Vehicle Re-identificatio
• IntraLoss: Further Margin via Gradient-Enhancing Term for Deep Face Recognition
• Learning and Interpreting Gravitational-Wave Features from CNNs with a Random Forest Approach
• Multi-Agent Semi-Siamese Training for Long-tail and Shallow Face Learning
• 3D Human Pose Estimation with Siamese Equivariant Embedding
• Learning Discriminative Features Via Weights-biased Softmax Loss
• GAM: Explainable Visual Similarity and Classification via Gradient Activation Maps
• Challenge of Spatial Cognition for Deep Learning
• Adma: A Flexible Loss Function for Neural Networks
• On the human evaluation of audio adversarial examples
• Deep Multi-task Multi-label CNN for Effective Facial Attribute Classification
• Directional Deep Embedding and Appearance Learning for Fast Video Object Segmentation
• Deep Metric Learning with Density Adaptivity
• On Learning Density Aware Embeddings
• Intelligent Frame Selection as a Privacy-Friendlier Alternative to Face Recognition
• AuthNet: A Deep Learning based Authentication Mechanism using Temporal Facial Feature Movements
• Structure-Aware Face Clustering on a Large-Scale Graph with $10^7$ Nodes
• Imbalance Robust Softmax for Deep Embeeding Learning
• Cosmetic-Aware Makeup Cleanser
• AMC-Loss: Angular Margin Contrastive Loss for Improved Explainability in Image Classification
• A Novel Perspective to Zero-shot Learning: Towards an Alignment of Manifold Structures via Semantic Feature Expansion
• Improved Super-Resolution Convolution Neural Network for Large Images
• Attribute Guided Sparse Tensor-Based Model for Person Re-Identification
• Occlusion-guided compact template learning for ensemble deep network-based pose-invariant face recognition
• Robustness Analysis of Face Obscuration
• Alpha-Net: Architecture, Models, and Applications
• ECML: An Ensemble Cascade Metric Learning Mechanism towards Face Verification
• Learn to Propagate Reliably on Noisy Affinity Graphs
• Signed Graph Metric Learning via Gershgorin Disc Perfect Alignment
• Profile to Frontal Face Recognition in the Wild Using Coupled Conditional GAN
• Person Re-identification Attack on Wearable Sensing
• FinBrain: When Finance Meets AI 2.0
• Representation Learning from Limited Educational Data with Crowdsourced Labels
• Face representation by deep learning: a linear encoding in a parameter space?
• Unknown Identity Rejection Loss: Utilizing Unlabeled Data for Face Recognition
• A Discriminative Learned CNN Embedding for Remote Sensing Image Scene Classification
• Adaptive Deep Metric Embeddings for Person Re-Identification under Occlusions
• Stop memorizing: A data-dependent regularization framework for intrinsic pattern learning
• Improved Deep Hashing with Soft Pairwise Similarity for Multi-label Image Retrieval
• An Alternative Practice of Tropical Convolution to Traditional Convolutional Neural Networks
• Identity Recognition in Intelligent Cars with Behavioral Data and LSTM-ResNet Classifier
• Learning discriminative features in sequence training without requiring framewise labelled data
• MSFD:Multi-Scale Receptive Field Face Detector
• DAIL: Dataset-Aware and Invariant Learning for Face Recognition
• Boosting Unconstrained Face Recognition with Auxiliary Unlabeled Data
• "Who is Driving around Me?" Unique Vehicle Instance Classification using Deep Neural Features