More Information Supervised Probabilistic Deep Face Embedding Learning
arXiv:2006.04518
Abstract
Researches using margin based comparison loss demonstrate the effectiveness of penalizing the distance between face feature and their corresponding class centers. Despite their popularity and excellent performance, they do not explicitly encourage the generic embedding learning for an open set recognition problem. In this paper, we analyse margin based softmax loss in probability view. With this perspective, we propose two general principles: 1) monotonic decreasing and 2) margin probability penalty, for designing new margin loss functions. Unlike methods optimized with single comparison metric, we provide a new perspective to treat open set face recognition as a problem of information transmission. And the generalization capability for face embedding is gained with more clean information. An auto-encoder architecture called Linear-Auto-TS-Encoder(LATSE) is proposed to corroborate this finding. Extensive experiments on several benchmarks demonstrate that LATSE help face embedding to gain more generalization capability and it boosted the single model performance with open training dataset to more than on MegaFace test.
References in corpus (20)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Distilling the Knowledge in a Neural Network
- Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
- MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
- Learning Face Representation from Scratch
- Recent Advances in Open Set Recognition: A Survey
- Energy-based Generative Adversarial Network
- Learning Deconvolution Network for Semantic Segmentation
- MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labels
- Training Deep Nets with Sublinear Memory Cost
- Large-Margin Softmax Loss for Convolutional Neural Networks
- RetinaFace: Single-stage Dense Face Localisation in the Wild
- SphereFace: Deep Hypersphere Embedding for Face Recognition
- Cross-Age LFW: A Database for Studying Cross-Age Face Recognition in Unconstrained Environments
- CosFace: Large Margin Cosine Loss for Deep Face Recognition
- How does Disagreement Help Generalization against Label Corruption?
- Support Vector Guided Softmax Loss for Face Recognition
- AdaCos: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations
- The Devil of Face Recognition is in the Noise
- Angular Visual Hardness