Deep Imbalanced Learning for Face Recognition and Attribute Prediction

TL;DR

This paper introduces Cluster-based Large Margin Local Embedding (CLMLE), a novel deep learning approach that improves face recognition and attribute prediction on imbalanced datasets by enforcing inter-cluster margins, leading to more balanced class boundaries.

Contribution

It proposes a new deep representation learning method that maintains inter-cluster margins on a hypersphere, enhancing performance on imbalanced face analysis tasks.

Findings

CLMLE outperforms existing methods in face recognition accuracy.

The approach effectively reduces class imbalance effects locally.

Significant improvements in face attribute prediction accuracy.

Abstract

Data for face analysis often exhibit highly-skewed class distribution, i.e., most data belong to a few majority classes, while the minority classes only contain a scarce amount of instances. To mitigate this issue, contemporary deep learning methods typically follow classic strategies such as class re-sampling or cost-sensitive training. In this paper, we conduct extensive and systematic experiments to validate the effectiveness of these classic schemes for representation learning on class-imbalanced data. We further demonstrate that more discriminative deep representation can be learned by enforcing a deep network to maintain inter-cluster margins both within and between classes. This tight constraint effectively reduces the class imbalance inherent in the local data neighborhood, thus carving much more balanced class boundaries locally. We show that it is easy to deploy angular margins between the cluster distributions on a hypersphere manifold. Such learned Cluster-based Large Margin Local Embedding (CLMLE), when combined with a simple k-nearest cluster algorithm, shows significant improvements in accuracy over existing methods on both face recognition and face attribute prediction tasks that exhibit imbalanced class distribution.