Neural Class-Specific Regression for face verification

TL;DR

This paper introduces a regression-based approach to class-specific face verification, enabling scalable and efficient training for large datasets using linear, kernel, and neural network methods.

Contribution

It generalizes previous kernel-based discriminant analysis into a regression framework, facilitating large-scale face verification with efficient training schemes.

Findings

Kernel-based methods perform well on small datasets.

Neural network methods scale effectively to large datasets.

Regression approach improves training efficiency for large-scale problems.

Abstract

Face verification is a problem approached in the literature mainly using nonlinear class-specific subspace learning techniques. While it has been shown that kernel-based Class-Specific Discriminant Analysis is able to provide excellent performance in small- and medium-scale face verification problems, its application in today's large-scale problems is difficult due to its training space and computational requirements. In this paper, generalizing our previous work on kernel-based class-specific discriminant analysis, we show that class-specific subspace learning can be cast as a regression problem. This allows us to derive linear, (reduced) kernel and neural network-based class-specific discriminant analysis methods using efficient batch and/or iterative training schemes, suited for large-scale learning problems. We test the performance of these methods in two datasets describing medium- and large-scale face verification problems.