Domain Adaptation with Soft-margin multiple feature-kernel learning beats Deep Learning for surveillance face recognition
This work addresses the challenge of low-performance face recognition in surveillance for security applications, offering a novel approach that outperforms deep learning in these specific conditions.
The paper tackles the problem of face recognition in surveillance scenarios with poor image quality by proposing a soft-margin multiple feature-kernel learning method with domain adaptation, achieving superior performance over state-of-the-art techniques on three real-world datasets.
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.