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 (23)

• Learning Deep Embeddings with Histogram Loss
• Deep Adaptive Feature Embedding with Local Sample Distributions for Person Re-identification
• Personalized Age Progression with Bi-level Aging Dictionary Learning
• Deep learning and face recognition: the state of the art
• Graph-RISE: Graph-Regularized Image Semantic Embedding
• AdaCos: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations
• ShrinkTeaNet: Million-scale Lightweight Face Recognition via Shrinking Teacher-Student Networks
• A Light Dual-Task Neural Network for Haze Removal
• Cross-Resolution Face Recognition via Prior-Aided Face Hallucination and Residual Knowledge Distillation
• Decorrelated Adversarial Learning for Age-Invariant Face Recognition
• Character-level Chinese Writer Identification using Path Signature Feature, DropStroke and Deep CNN
• Signal-to-Noise Ratio: A Robust Distance Metric for Deep Metric Learning
• Multi-Prototype Networks for Unconstrained Set-based Face Recognition
• DropSample: A New Training Method to Enhance Deep Convolutional Neural Networks for Large-Scale Unconstrained Handwritten Chinese Character Recognition
• Triplet Distillation for Deep Face Recognition
• Style Transfer Applied to Face Liveness Detection with User-Centered Models
• Learning Semantic Vector Representations of Source Code via a Siamese Neural Network
• MassFace: an efficient implementation using triplet loss for face recognition
• Learning Discriminative Features Via Weights-biased Softmax Loss
• Learning discriminative features in sequence training without requiring framewise labelled data
• MSFD:Multi-Scale Receptive Field Face Detector
• Occlusion-guided compact template learning for ensemble deep network-based pose-invariant face recognition
• On Learning Density Aware Embeddings