von Mises-Fisher Mixture Model-based Deep learning: Application to Face Verification

Abstract

A number of pattern recognition tasks, \textit{e.g.}, face verification, can be boiled down to classification or clustering of unit length directional feature vectors whose distance can be simply computed by their angle. In this paper, we propose the von Mises-Fisher (vMF) mixture model as the theoretical foundation for an effective deep-learning of such directional features and derive a novel vMF Mixture Loss and its corresponding vMF deep features. The proposed vMF feature learning achieves the characteristics of discriminative learning, \textit{i.e.}, compacting the instances of the same class while increasing the distance of instances from different classes. Moreover, it subsumes a number of popular loss functions as well as an effective method in deep learning, namely normalization. We conduct extensive experiments on face verification using 4 different challenging face datasets, \textit{i.e.}, LFW, YouTube faces, CACD and IJB-A. Results show the effectiveness and excellent generalization ability of the proposed approach as it achieves state-of-the-art results on the LFW, YouTube faces and CACD datasets and competitive results on the IJB-A dataset.

Under review

References in corpus (24)

• Auto-Encoding Variational Bayes
• Learning Face Representation from Scratch
• DeepID3: Face Recognition with Very Deep Neural Networks
• Autoencoding beyond pixels using a learned similarity metric
• Robust Face Recognition via Multimodal Deep Face Representation
• Large-Margin Softmax Loss for Convolutional Neural Networks
• SphereFace: Deep Hypersphere Embedding for Face Recognition
• Recent Advances in Convolutional Neural Networks
• Cross-Age LFW: A Database for Studying Cross-Age Face Recognition in Unconstrained Environments
• Targeting Ultimate Accuracy: Face Recognition via Deep Embedding
• Naive-Deep Face Recognition: Touching the Limit of LFW Benchmark or Not?
• MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition
• A Light CNN for Deep Face Representation with Noisy Labels
• Do We Really Need to Collect Millions of Faces for Effective Face Recognition?
• Range Loss for Deep Face Recognition with Long-tail
• Neural Aggregation Network for Video Face Recognition
• Triplet Similarity Embedding for Face Verification
• Joint Color-Spatial-Directional clustering and Region Merging (JCSD-RM) for unsupervised RGB-D image segmentation
• An All-In-One Convolutional Neural Network for Face Analysis
• Cosine Normalization: Using Cosine Similarity Instead of Dot Product in Neural Networks
• DeepVisage: Making face recognition simple yet with powerful generalization skills
• Automatic Face Image Quality Prediction
• VIPLFaceNet: An Open Source Deep Face Recognition SDK
• Improving Shadow Suppression for Illumination Robust Face Recognition

Cited by in corpus (19)

• Deep Face Recognition: A Survey
• CosFace: Large Margin Cosine Loss for Deep Face Recognition
• Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere
• Hyperspherical Variational Auto-Encoders
• Hyperspherical Prototype Networks
• Riemannian Continuous Normalizing Flows
• Spherical Latent Spaces for Stable Variational Autoencoders
• Regular Polytope Networks
• Visual Concepts and Compositional Voting
• Generalized Inter-class Loss for Gait Recognition
• Fix Your Features: Stationary and Maximally Discriminative Embeddings using Regular Polytope (Fixed Classifier) Networks
• A Bayesian Approach to OOD Robustness in Image Classification
• Discriminative Few-Shot Learning Based on Directional Statistics
• Compressive Visual Representations
• Noise-Resistant Deep Metric Learning with Probabilistic Instance Filtering
• Directional Deep Embedding and Appearance Learning for Fast Video Object Segmentation
• On The Distribution of Penultimate Activations of Classification Networks
• Profile to Frontal Face Recognition in the Wild Using Coupled Conditional GAN
• Learning Representations on the Unit Sphere: Investigating Angular Gaussian and von Mises-Fisher Distributions for Online Continual Learning