Microsoft’s depth-sensing Kinect device has one major shortcoming: the inability to work in sunlight or other bright light.
Now, according to a new study presented Monday at SIGGRAPH 2015, the International Conference on Computer Graphics and Interactive Techniques, researchers have developed a digital camera device capable of addressing that shortfall.
According to the study team, the device works by collecting only the parts of light the camera actually needs. To do this, the scientists developed a mathematical simulation to help program these devices so the camera and its light source interact efficiently, removing extraneous light that would usually wash out the signals required to recognize a scene’s shapes.
“We have a way of choosing the light rays we want to capture and only those rays,” team member Srinivasa Narasimhan, an associate professor of robotics at Carnegie Mellon University, said in a press release. “We don’t need new image-processing algorithms and we don’t need extra processing to eliminate the noise, because we don’t collect the noise. This is all done by the sensor.”
The team was able to develop prototype founded on this simulation by syncing a laser projector with a typical rolling-shutter camera – the kind of camera found in most smartphones. This allows the camera to detect light solely from points being lit up by the laser as it scans the scene.
Works in any lighting setup
The novel system allows the camera to operate under a range of lighting setups. For example, it can tell the shape of a light bulb that has been turned on, or see through smoke. As a side benefit, the device is also extremely energy efficient.
The research team said their technology could find uses in a variety of settings, including medical imaging, assessment of shiny metallic parts, and robotic vision. Additionally, it might be readily incorporated into smartphones.
Team member Kyros Kutulakos, University of Toronto professor of computer science, explained that the system works by basically overpowering sunlight.
“Even though we’re not sending a huge amount of photons, at short time scales, we’re sending a lot more energy to that spot than the energy sent by the sun,” he said. The key is to be able to record just the light from that spot as it is lit up, as opposed to try to pick out the spot from a bright scene.
(Image: The prototype is able to capture the detail of a lightbulb that would otherwise just be a blinding blob of light to a normal camera. Credit: Carnegie Mellon University)