Measuring objects in 3D using only a camera and projector
A group at AIST has developed technology for measuring objects in 3D, using only a camera and projector.
With this new method, patterned light is projected onto the object, and the pattern images captured by the camera are processed to measure it's 3D shape. Because the 3D shape can be measured from a single image, this method can also be used with a high-speed camera.
"If you look carefully, you can see that each line is wavy. The wavy line patterns are carefully designed, so the pattern from the projector can be recognized from the camera image. This makes it possible to use the projector and the camera for triangulation, so we can know the shape at that instant."
This method provides more measurement positions than conventional motion capture. Consequently, it can measure wrinkles in clothing, or fine details of hands. The method is precise, with a tolerance of just 1-2 mm. So, it's expected to be useful in fields where shapes couldn't be measured before because they change too fast.
"For example, there's been research on using the body to control things like multimedia, virtual reality, and games, and this system can be used as part of technologies like that. Regarding applications, we think data obtained by this measurement method could be used to analyze how athletes move."
One feature of this system is that as long as the patterns can be captured, it can be applied to different devices, too. For example, if a stereo microscope is used to create a system that projects patterns in a very small area, it would be possible to measure shapes from microscope images.
"Currently, we've stopped at the stage of making measurements. But we're also thinking about how to use the measured data. We'd like to work on applications to sports science and materials analysis. We'd also like to keep increasing the range of things that can be measured. We think this method could be used to determine how much buildings would shake in an earthquake. In the future, we hope it will be possible to measure, for example, the movements of an athlete in the middle of a stadium, from a long way away. If we can do that, we think this method would be great for making 3D videos as well."