Army: Facial recognition, even in the dark

ABERDEEN PROVING GROUND, Md. Researchers at the U.S. Army Research Laboratory have developed a technique that uses artificial intelligence and machine learning to produce a recognizable image from a thermal image of a person's face captured in low-light or nighttime conditions. Such a technique could lead to enhanced real-time biometrics and post-mission forensic analysis for covert military nighttime operations.

A conceptual illustration for thermal-to-visible synthesis for interoperability with existing visible-based facial recognition systems. (Courtesy Eric Proctor, William Parks, and Benjamin S. Riggan)

The newly developed approach -- which leverages advanced domain adaptation techniques based on deep neural networks has two key parts: a nonlinear regression model that maps a given thermal image into a corresponding visible latent representation, plus an optimization problem that projects the latent projection back into the image space.

The optimization problem for synthesizing an image attempts to jointly preserve the shape of the entire face and appearance of the local reference points. Using the synthesized thermal-to-visible imagery and existing visible gallery imagery, the researchers performed face verification experiments using a common open source deep neural network architecture for face recognition; such architecture is designed solely for visible-based face recognition. The most surprising result: This approach achieved better verification performance than a generative adversarial network-based approach, which previously showed photorealistic properties.

The motivation for this "see in the dark" technology -- developed by Drs. Benjamin S. Riggan, Nathaniel J. Short, and Shuowen "Sean" Hu, from the U.S. Army Research Laboratory -- is to enhance both automatic and human-matching capabilities. "This technology enables matching between thermal face images and existing biometric face databases/watch lists that only contain visible face imagery," said Riggan, a research scientist. "The technology provides a way for humans to visually compare visible and thermal facial imagery through thermal-to-visible face synthesis."

Riggan said that he and his fellow researchers will continue to extend this project under the sponsorship of the Defense Forensics and Biometrics Agency as they attempt to develop a robust nighttime face-recognition capability for military use.