Researchers have developed a machine learning technique that produces a visible face image from a thermal image of a person's face captured in the dark.
It's hoped that the development of this AI technique could lead to enhanced real-time biometrics and post-mission forensic analysis for covert night-time operations for the US army.
The US Army Research Laboratory, who is behind this finding, said in a paper that the ability to perform automatic face recognition at night-time using such thermal cameras is beneficial for informing a soldier that an individual is someone of interest, like someone who may be on a watch list. It could also help 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 research scientist Dr. Benjamin S. Riggan, who developed the tech. "The technology provides a way for humans to visually compare visible and thermal facial imagery through thermal-to-visible face synthesis."
He explained that under night-time and low-light conditions, there is currently insufficient light for a conventional camera to capture facial imagery for recognition without active illumination such as a flash or spotlight, which would give away the position of such surveillance cameras; however, thermal cameras that capture the heat signature naturally emanating from living skin tissue are ideal for such conditions.
"When using thermal cameras to capture facial imagery, the main challenge is that the captured thermal image must be matched against a watch list or gallery that only contains conventional visible imagery from known persons of interest," Riggan added.
"Therefore, the problem becomes what is referred to as cross-spectrum, or heterogeneous, face recognition. In this case, facial probe imagery acquired in one modality is matched against a gallery database acquired using a different imaging modality."
Riggan said he and his colleagues will continue to extend this research under the sponsorship of the Defense Forensics and Biometrics Agency to develop a robust night-time face recognition capability for soldiers.
RTX 280 Ti will come with 11GB of fast GDDR6 video RAM with a 352-bit memory bus offering 616Gbps
The scale of jobs lost to automation will be at least as large as those in the first three industrial revolutions
Latest Tesla news: Tesla stock price tanks amid reports of 'widening probe' by SEC and claims the base Model 3 loses money
SEC 'probe' takes its toll on Tesla as new research suggests that Tesla loses $6,000 on every $35,000 Model 3
10nm Cannon Lake Core i3-8121U CPUs make a rare outing with Intel's NUC mini PC