Feature Transfer Learning for Deep Face Recognition with Under-Represented Data

Abstract

Despite the large volume of face recognition datasets, there is a significant portion of subjects, of which the samples are insufficient and thus under-represented. Ignoring such significant portion results in insufficient training data. Training with under-represented data leads to biased classifiers in conventionally-trained deep networks. In this paper, we propose a center-based feature transfer framework to augment the feature space of under-represented subjects from the regular subjects that have sufficiently diverse samples. A Gaussian prior of the variance is assumed across all subjects and the variance from regular ones are transferred to the under-represented ones. This encourages the under-represented distribution to be closer to the regular distribution. Further, an alternating training regimen is proposed to simultaneously achieve less biased classifiers and a more discriminative feature representation. We conduct ablative study to mimic the under-represented datasets by varying the portion of under-represented classes on the MS-Celeb-1M dataset. Advantageous results on LFW, IJB-A and MS-Celeb-1M demonstrate the effectiveness of our feature transfer and training strategy, compared to both general baselines and state-of-the-art methods. Moreover, our feature transfer successfully presents smooth visual interpolation, which conducts disentanglement to preserve identity of a class while augmenting its feature space with non-identity variations such as pose and lighting.

arXiv commentCVPR2019

References in corpus (14)

• SMOTE: Synthetic Minority Over-sampling Technique
• Prototypical Networks for Few-shot Learning
• Matching Networks for One Shot Learning
• Large-Margin Softmax Loss for Convolutional Neural Networks
• SphereFace: Deep Hypersphere Embedding for Face Recognition
• CosFace: Large Margin Cosine Loss for Deep Face Recognition
• MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition
• A Light CNN for Deep Face Representation with Noisy Labels
• The Devil is in the Tails: Fine-grained Classification in the Wild
• Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and Identity Preserving Frontal View Synthesis
• One-shot Face Recognition by Promoting Underrepresented Classes
• Range Loss for Deep Face Recognition with Long-tail
• Towards Large-Pose Face Frontalization in the Wild
• Reconstruction-Based Disentanglement for Pose-invariant Face Recognition

Cited by in corpus (10)

• Deep Face Recognition: A Survey
• CReST: A Class-Rebalancing Self-Training Framework for Imbalanced Semi-Supervised Learning
• Striking the Right Balance with Uncertainty
• Deep Imbalanced Learning for Face Recognition and Attribute Prediction
• Max-margin Class Imbalanced Learning with Gaussian Affinity
• Feature Space Augmentation for Long-Tailed Data
• Multi-Level Analyzation of Imbalance to Resolve Non-IID-Ness in Federated Learning
• Resembled Generative Adversarial Networks: Two Domains with Similar Attributes
• Multi-Agent Semi-Siamese Training for Long-tail and Shallow Face Learning
• Online Continual Learning in Image Classification: An Empirical Survey