Crystal Loss and Quality Pooling for Unconstrained Face Verification and Recognition

Abstract

In recent years, the performance of face verification and recognition systems based on deep convolutional neural networks (DCNNs) has significantly improved. A typical pipeline for face verification includes training a deep network for subject classification with softmax loss, using the penultimate layer output as the feature descriptor, and generating a cosine similarity score given a pair of face images or videos. The softmax loss function does not optimize the features to have higher similarity score for positive pairs and lower similarity score for negative pairs, which leads to a performance gap. In this paper, we propose a new loss function, called Crystal Loss, that restricts the features to lie on a hypersphere of a fixed radius. The loss can be easily implemented using existing deep learning frameworks. We show that integrating this simple step in the training pipeline significantly improves the performance of face verification and recognition systems. We achieve state-of-the-art performance for face verification and recognition on challenging LFW, IJB-A, IJB-B and IJB-C datasets over a large range of false alarm rates (10-1 to 10-7).

arXiv commentPreviously portions of this work appeared in arXiv:1703.09507, which was a conference version. This version is an extended journal version of it

References in corpus (8)

• Caffe: Convolutional Architecture for Fast Feature Embedding
• SphereFace: Deep Hypersphere Embedding for Face Recognition
• Aggregated Residual Transformations for Deep Neural Networks
• Targeting Ultimate Accuracy: Face Recognition via Deep Embedding
• Face Search at Scale: 80 Million Gallery
• Neural Aggregation Network for Video Face Recognition
• A Good Practice Towards Top Performance of Face Recognition: Transferred Deep Feature Fusion
• DeepVisage: Making face recognition simple yet with powerful generalization skills

Cited by in corpus (17)

• SFace: Sigmoid-Constrained Hypersphere Loss for Robust Face Recognition
• AdaCos: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations
• WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognition
• Recurrent Embedding Aggregation Network for Video Face Recognition
• Video Face Recognition: Component-wise Feature Aggregation Network (C-FAN)
• Deep Face Recognition Model Compression via Knowledge Transfer and Distillation
• Single Unit Status in Deep Convolutional Neural Network Codes for Face Identification: Sparseness Redefined
• An Automatic System for Unconstrained Video-Based Face Recognition
• NPT-Loss: A Metric Loss with Implicit Mining for Face Recognition
• Deep Regionlets: Blended Representation and Deep Learning for Generic Object Detection
• A Fast and Accurate System for Face Detection, Identification, and Verification
• EQFace: A Simple Explicit Quality Network for Face Recognition
• Multi-Facet Recommender Networks with Spherical Optimization
• On measuring the iconicity of a face
• Introducing the DOME Activation Functions
• Self-attention aggregation network for video face representation and recognition
• Evaluating the Effectiveness of Automated Identity Masking (AIM) Methods with Human Perception and a Deep Convolutional Neural Network (CNN)