Editorial Introduction to the Special Issue on Biometrics at a Distance in the Deep Learning Era

Manuel J. Marín-Jiménez, Shiqi Yu, Yasushi Makihara, Vishal M. Patel, Maneet Singh, Maria De Marsico

Abstract: A total of 36 papers were submitted to this SI, where 14 papers were accepted after a rigorous review process. Three papers in the final collection aim at improving the performance of people re-identification systems, including the use of visible and infrared images, proposing a new retrieval and verification loss function, and suggesting the fusion of visual appearance and soft biometrics. Five works focus on faces, addressing diverse issues such as face image completion or dealing with low-resolution faces, even those affected by turbulences. Three works deal with other human body parts, such as the eye, the palm, and the finger knuckle. Two papers use human motion as a biometric feature. Finally, a new dataset of pedestrians is introduced for soft biometrics at a distance. A more detailed introduction to the SI topics follows.