Domain Adaptation with Soft-margin multiple feature-kernel learning beats Deep Learning for surveillance face recognition

TL;DR

This paper introduces a novel soft-margin multiple feature-kernel learning approach combined with domain adaptation, significantly improving face recognition performance in challenging surveillance conditions over deep learning methods.

Contribution

It presents a new soft-margin learning method for multiple feature-kernel combinations with domain adaptation, outperforming recent state-of-the-art techniques in surveillance face recognition.

Findings

Outperforms recent state-of-the-art methods on real-world datasets

Effective in low contrast, noisy, and poorly illuminated conditions

Demonstrates robustness of the proposed approach in surveillance scenarios

Abstract

Face recognition (FR) is the most preferred mode for biometric-based surveillance, due to its passive nature of detecting subjects, amongst all different types of biometric traits. FR under surveillance scenario does not give satisfactory performance due to low contrast, noise and poor illumination conditions on probes, as compared to the training samples. A state-of-the-art technology, Deep Learning, even fails to perform well in these scenarios. We propose a novel soft-margin based learning method for multiple feature-kernel combinations, followed by feature transformed using Domain Adaptation, which outperforms many recent state-of-the-art techniques, when tested using three real-world surveillance face datasets.