Hacker News Reader

Comment by gavanwoolery | Hacker News Reader

Similar products have existed for a long time, but mostly "in the lab." Nonetheless, it is cool, although I suspect there would be a lot of accidental gesturing -- there are many ways to prevent accidental gestures, but all at the expense of usability.

View on HN

Who cares? Put that thing on with the Oculus Rift. You'll want to capture the full range of motion, because you'll determine gestures by their interactions with the virtual environment.

I wish I could upvote you more than once. People will look at this video and think "oh, they've solved the big problem!" when in fact they've solved half of the big problem. Distinguishing between intentional and unintentional gesture has been the academic elephant in the room for many years now, and no one has come up with a good solution yet.

Their FAQ states they've resolved this issue by leveraging unnatural gestures. There's no way to tell if their solution is any good but at least this should give the public an opportunity to decide which applications the tech is ready for and which it isn't. Things that tend to be difficult to define "in the lab."

For one off gestures, you could also very well have a quick "trigger" gesture that enables the device, and gets it going. When you are done, use trigger again,

I'd be interested in other references.

I'm thinking it would be easier to create a distinguishing 'attention' gesture than it would be with other things. The finger exercise thing comes to mind. Also much easier to do than a voice activated thing like calling out to Siri in a restaurant or something similar.

They show the band fairly far up the arm. Presumably to get a good set of distinguishing signals. But I wonder how hard it would be with enough DSP to build this into a form fitting wrap around watch band? The combination of gesture articulation and appli-wrist-watch is a compelling one.

Reply on news.ycombinator.com