18 ms·

Main problem is there is that as all information used by the leap is continuous in nature there is essentially no (forseeable) way to unequivocally convey the i

Main problem is there is that as all information used by the leap is continuous in nature there is essentially no (forseeable) way to unequivocally convey the intended start and end of a gesture. I personally imagine an interface not unlike a wand - people might carry around a pointing stick of some kind, with a button built into the handle to confirm gestures either on or off (at least until we can start doing crazy things like gauging intent - what if the computer only responded to your motions if it knew you were looking at it?). Probably a terrible idea.

newishuser 14y ago
Bend your index finger to touch your thumb. Something as trivial as that could mean "Don't treat this as an interaction".
- malandrew 14y ago
  Even better: initiate a gesture with a closed fist (no fingers detected), extend appropriate fingers for gesture, reform a fist to end the gesture.
  - Macs75 14y ago
    If is so precise I think gesture with less movements are the best. To have the same value of using a mouse for me the best could be to have a finger pointing/moving cursor and to have the middle finger touching the thumb and closing a circle during the gesture draw. When you open the "circle" ( slightly detach the middle finger from the thumb) the gesture is complete and could be interpreted. In the same way you could simulate single click, double clicks using always the thumb as a button, finally you could use the other fingers on the thumb to have a right click or more buttons. Maybe it's possible to simulate the wheel sliding the finger on the surface of the thumb. If you try those movements you will understand how is easy and natural to adopt and adapt to them. Imagine to drag and drop a windows on a desktop.. point the bar close the circle to "click" then drag the window and last open the circle to release. This way it's also easy to interface normal gesture build on drawing like the existing ones for mouses and mix also movement of the other hand to obtain more complex stuff like rotations,zoom and other complex manipulations if needed.
lifeisstillgood 14y ago
It seems this is useful for interpreting sign language - and may be another GUI for talking to a computer - but if you have sign language you can type ... Although I can see the value in accurate gesturing in a room to say turn off the lights.
gghootch 14y ago
Try to move away from the PC analogy. Imagine a 30" display, running iPad applications. Your input consists of the exact same gestures, but now in the air instead of touching the screen. Painting, drum-machines, games and many more applications come to mind. DJs moving from turning knobs on impenetrable devices to conducting a live mix on a huge ass screen? Yes, please.

newishuser 14y ago

Bend your index finger to touch your thumb. Something as trivial as that could mean "Don't treat this as an interaction".

malandrew 14y ago
Even better: initiate a gesture with a closed fist (no fingers detected), extend appropriate fingers for gesture, reform a fist to end the gesture.
- Macs75 14y ago
  If is so precise I think gesture with less movements are the best. To have the same value of using a mouse for me the best could be to have a finger pointing/moving cursor and to have the middle finger touching the thumb and closing a circle during the gesture draw. When you open the "circle" ( slightly detach the middle finger from the thumb) the gesture is complete and could be interpreted. In the same way you could simulate single click, double clicks using always the thumb as a button, finally you could use the other fingers on the thumb to have a right click or more buttons. Maybe it's possible to simulate the wheel sliding the finger on the surface of the thumb. If you try those movements you will understand how is easy and natural to adopt and adapt to them. Imagine to drag and drop a windows on a desktop.. point the bar close the circle to "click" then drag the window and last open the circle to release. This way it's also easy to interface normal gesture build on drawing like the existing ones for mouses and mix also movement of the other hand to obtain more complex stuff like rotations,zoom and other complex manipulations if needed.

lifeisstillgood 14y ago

It seems this is useful for interpreting sign language - and may be another GUI for talking to a computer - but if you have sign language you can type ... Although I can see the value in accurate gesturing in a room to say turn off the lights.

gghootch 14y ago

Try to move away from the PC analogy. Imagine a 30" display, running iPad applications. Your input consists of the exact same gestures, but now in the air instead of touching the screen. Painting, drum-machines, games and many more applications come to mind. DJs moving from turning knobs on impenetrable devices to conducting a live mix on a huge ass screen? Yes, please.