Posted by
Microsoft Research Blog
Introducing Aurora: The first large-scale foundation model of the atmosphere
Aurora, a new AI foundation model from Microsoft Research, can transform our ability to predict and mitigate extreme weather events and the effects of climate change by enabling faster and more accurate weather forecasts than ever before.
Now, it’s time for KinÊtre (opens in new tab), the latest Microsoft Research project that uses a Kinect depth camera to provide novel functionality: Among other animations, it can make chairs dance.
No, your eyes didn’t deceive: dancing chairs. If that sounds like fun, you’ve come to the right place.
KinÊtre, which can animate all sorts of inanimate objects, will be presented as a talk by Jiawen Chen (opens in new tab) during SIGGRAPH 2012 (opens in new tab), being held at the Los Angeles Convention Center from Aug. 5 to 9. Perhaps the best introduction to the project is this informative video:
“The idea is that we want to let you bring the world around you to life,” Chen says. “We want to let you use Kinect for Windows and a PC and take arbitrary household objects and make them move like a cartoon character.”
The process to engage is deceptively simple. You use a consumer depth camera to scan an item such as a chair, a desk lamp, a bookcase, or a stepladder. Then you use the same camera to track your body and align your virtual limbs to the geometry of the item. With a word—“Possess!”—your virtual limbs are attached to the item. When you jump, the bookcase jumps. When you dance, the chair dances.
And, using KinÊtre—a portmanteau that combines “Kinect” with the French verb for “to be”—anyone can create playful, 3-D models.
“We started with KinectFusion, but it scans static objects,” Chen says. “We want to bring things to life. We were thinking, ‘How would you potentially capture a moving object?’
“As a first step, we figured, ‘Why don’t we just capture a static object and then figure out an easy way to make it move?’”
Having done so, the researchers then began to investigate potential scenarios for use.
“When we started this, we were thinking of using it as a more effective way of doing set dressing and prop placement in movies for a preview,” Chen recalls. “Studios have large collections of shapes, and it’s pretty tedious to move them into place exactly. We wanted to be able to quickly walk around and grab things and twist them around. Then we realized we can do many more fun things.”
Other usage possibilities for KinÊtre include storytelling by children—the technology is sufficiently intuitive and accessible for kids to use—and gaming.
At the core of the technology is non-linear numerical optimization, enabling what Chen calls “a fairly robust, very easy method to make arbitrary things come to life.”
Still, the best way to understand KinÊtre is simply to see it in action. (Watch that video!) Chen gleefully recalls the reaction he got in March while demonstrating its capabilities during TechFest 2012 (opens in new tab).
“The best fun was seeing tons of Microsoft employees realize how fun it is,” he smiles. “Explaining it to them didn’t make too much sense. You have to experience it.”