Range Loss for Deep Face Recognition with Long-tail

Abstract

Convolutional neural networks have achieved great improvement on face recognition in recent years because of its extraordinary ability in learning discriminative features of people with different identities. To train such a well-designed deep network, tremendous amounts of data is indispensable. Long tail distribution specifically refers to the fact that a small number of generic entities appear frequently while other objects far less existing. Considering the existence of long tail distribution of the real world data, large but uniform distributed data are usually hard to retrieve. Empirical experiences and analysis show that classes with more samples will pose greater impact on the feature learning process and inversely cripple the whole models feature extracting ability on tail part data. Contrary to most of the existing works that alleviate this problem by simply cutting the tailed data for uniform distributions across the classes, this paper proposes a new loss function called range loss to effectively utilize the whole long tailed data in training process. More specifically, range loss is designed to reduce overall intra-personal variations while enlarging inter-personal differences within one mini-batch simultaneously when facing even extremely unbalanced data. The optimization objective of range loss is the $k$ greatest range's harmonic mean values in one class and the shortest inter-class distance within one batch. Extensive experiments on two famous and challenging face recognition benchmarks (Labeled Faces in the Wild (LFW) and YouTube Faces (YTF) not only demonstrate the effectiveness of the proposed approach in overcoming the long tail effect but also show the good generalization ability of the proposed approach.

9 pages, 5 figures, Submitted to CVPR, 2017

References in corpus (7)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Learning Face Representation from Scratch
• Going Deeper with Convolutions
• Object Detectors Emerge in Deep Scene CNNs
• Naive-Deep Face Recognition: Touching the Limit of LFW Benchmark or Not?
• MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition

Cited by in corpus (8)

• von Mises-Fisher Mixture Model-based Deep learning: Application to Face Verification
• Regularizing Class-wise Predictions via Self-knowledge Distillation
• Neural Architecture Search for Deep Face Recognition
• Generative One-Shot Face Recognition
• When 3D-Aided 2D Face Recognition Meets Deep Learning: An extended UR2D for Pose-Invariant Face Recognition
• Side Information for Face Completion: a Robust PCA Approach
• Imbalance Robust Softmax for Deep Embeeding Learning
• Unknown Identity Rejection Loss: Utilizing Unlabeled Data for Face Recognition