US military is developing a way to see through walls
DARPA has tapped RTX BBN Technologies to develop algorithms and tools that reconstruct the hidden geometry or infer the interior composition of man-made objects from distances approaching a kilometer.
This new project, called XENA, is all about using clever algorithms to piece together what's hidden behind barriers, even when the raw data is blurry or incomplete.
XENA can create images from minimal data inputs
Instead of needing perfect images, XENA uses smart math and pattern recognition to work with whatever info it gets—like combining a bunch of fuzzy photos into a clear picture; sources differ on whether this avoids large training datasets or relies on many low-quality simulated samples.
It can find hidden objects using far fewer signals than traditional scanners.
The technology could help in critical situations
BBN and Georgia Tech are building and testing this in the US, aiming to help commanders see what's behind closed doors when getting close isn't safe.
As BBN's Joshua Fasching puts it, "We are developing algorithms that turn a small number of grainy snapshots into enough detail for decision-makers to act...", which could change how emergencies or threats are handled—potentially reducing the need for risky close-ups.