News
News Categories

Microsoft dispenses with input peripherals with Handpose tracking software

By Koh Wanzi - on 20 Apr 2015, 11:35am

Microsoft dispenses with input peripherals with Handpose tracking software

Microsoft researchers used a combination of machine learning and 3D modeling to enable Handpose to track the small subtle movements of fingers. (Image Source: Microsoft)

In a time when companies like Leap Motion and Oculus Rift are giving us novel new experiences and ways to interact with our devices, it can appear awfully dated that physical input devices like mice, keyboards and controllers are still the primary way with which we interact with our machines.

But the latest system out of Microsoft Research holds great promise for the future of virtual reality and device interactions. Handpose is a system that purports to be able to track detailed hand motions, including the subtle positioning and twitches of our fingers.

It uses a camera to track hand movements, and while this is nothing new, Handpose stands out because of its ability to track a wider range of movements and accommodate more natural interactions with it. For instance, users will be able to stand up and move around a room while the camera tracks their movement and the smaller, subtler motions of their hands and fingers in real-time.

Because of the small size of our hands and fingers in relation to the rest of our body and the fact that fingers can often be hidden from the camera, it is difficult to precisely track and interpret hand motions. To achieve the higher level of fine motion tracking, researchers combined machine learning with 3D hand-modeling to help the computer understand and make sense of complex hand movements.

Handpose will work with a basic Kinect system, which should help boost adoption rates if it eventually finds its way onto the consumer market. Of course, the system is still in its research stage, but it is already teasing new possibilities for the future of virtual reality, gaming and even artificial intelligence (AI).

Microsoft says that Handpose could see a diverse range of uses, from law enforcement officials directing robots into dangerous situations to office workers who simply want to sort through documents with a few flips of the wrist. It could even help people with disabilities communicate by tracking the motions of sign language and translating them into the spoken word. And it could potentially offer gamers finer control over their actions in game, as opposed to manipulating their avatar with a clunky controller.

Furthermore, AI’s ability to understand hand motions could get a boost from the technology behind Handpose, making it easier for us to teach or instruct robots to perform specific tasks.

The more technically inclined among you can view the full paper for Handpose here, which will be presented at the CHI conference on human-computer interaction in Seoul later this year.

Source: Microsoft

Join HWZ's Telegram channel here and catch all the latest tech news!
Our articles may contain affiliate links. If you buy through these links, we may earn a small commission.