First responders and our men and women in uniform have dangerous jobs, so being able to see what’s around the corner before they get there is potentially lifesaving. Nvidia has done just that by creating something called the Virtual Eye in collaboration with a team at the Defense Advanced Research Projects Agency (DARPA).
Virtual Eye uses 3D and virtual reality along with a set of cameras to create an environment where first responders can take a virtual walk through an environment before sending humans in. Nvidia and DARPA call the platform “camera agnostic,” meaning any camera works to create the VR environment.
The possibilities are endless. A firefighter could position cameras in two parts of a room that’s on fire, and use the 3D images they create to plot a safe path through the flames, as well as locate any potential victims. On the battlefield, soldiers could look for potential ambushes, booby traps, or explosives that might put them in danger.
“Understanding what we see is critical to making the right decisions in the battlefield,” DARPA’s Virtual eye program director Trung Tran said. “We can create a 3D image by looking at the differences between two images, understanding the differences and fusing them together.”
Nvidia’s role in the project is to provide the GPUs that power the experience. Tran’s team says it selected Nvidia because the GPUs can fit into a laptop, and are powerful enough to enable a real-time experience. While sports broadcasts already use cameras to stitch together a 3D image, it’s not in real time, so its use in a first responder or military environment isn’t that practical.
While the Virtual Eye is already available for use, it can only currently support two cameras. Tran says his team is working to enable the platform to support several cameras, with a goal of having a five-camera system viable by early 2017.