Heads Turn as Microsoft Shows Off 3-D Scanning Techniques
Scanning a three-dimensional image has been possible for quite a while now. The challenge has been doing so quickly, easily and inexpensively.
At Microsoft’s TechFest this week, the software maker is showing off several different methods, each of which is capable of producing a 3-D scan of a person or object in pretty short order.
While 3-D image capture is just one of dozens of areas being pursued by Microsoft’s 800-person research team, the company chose to highlight its work in that area as part of the public portion of the its internal science fair.
Having multiple projects — even competing efforts — in the same area is part of the culture of Microsoft Research, says corporate VP Peter Lee, who heads all of the company’s U.S. labs.
While 3-D scanning may not be a critical need for many users today, it is important to the future of gaming and augmented reality, along with business uses. Besides, Lee says, access to an easy means of scanning makes that MakerBot 3-D printer he has at home a lot more useful.
Three different means for creating 3-D scans were shown off at TechFest, but the lab is looking at even more options, Lee said. At the highest-quality end, the company demonstrated an impressive scan using a small green screen combined with an array of six digital cameras.
Two other research projects, meanwhile, produced pretty decent scans using even more basic gear.
The first method used the depth-sensing camera on the Kinect for Windows sensor, while another effort used only a custom app on a Windows Phone to gather the needed images.
Lee said that, internally, Microsoft seems to like the Kinect option, and plans to include the feature in the next version of the Kinect for Windows software toolkit for developers.
“We’ll see what our customers do,” he said.
Here’s a video showing all three projects in action: