Quadratic Projection Based Feature Extraction with Its Application to Biometric Recognition

TL;DR

This paper introduces a quadratic projection feature extraction method using a scalable semidefinite programming approach, demonstrating improved biometric recognition across face, palmprint, and ear datasets.

Contribution

It proposes a novel quadratic matrix learning framework with an efficient DualQML algorithm for high-dimensional data, enhancing biometric recognition performance.

Findings

Superior recognition accuracy on face, palmprint, and ear datasets

Effective scalability of the quadratic matrix learning method

Outperforms current state-of-the-art algorithms in biometric tasks

Abstract

This paper presents a novel quadratic projection based feature extraction framework, where a set of quadratic matrices is learned to distinguish each class from all other classes. We formulate quadratic matrix learning (QML) as a standard semidefinite programming (SDP) problem. However, the con- ventional interior-point SDP solvers do not scale well to the problem of QML for high-dimensional data. To solve the scalability of QML, we develop an efficient algorithm, termed DualQML, based on the Lagrange duality theory, to extract nonlinear features. To evaluate the feasibility and effectiveness of the proposed framework, we conduct extensive experiments on biometric recognition. Experimental results on three representative biometric recogni- tion tasks, including face, palmprint, and ear recognition, demonstrate the superiority of the DualQML-based feature extraction algorithm compared to the current state-of-the-art algorithms