Leap Motion. Impressions

First impressions.

After I connected the device for the first time and tried out test programs on my laptop, I was absolutely delighted! There was a feeling that here it is, the future, a little more, and the world will turn into something like this video .

Well, indeed, what other impressions of computer control "in the air" could be! Even the kinect was not as interesting as this device.

Inspired, I immediately began to figure out how to do something for myself!

I took the first simple project I got and decided to try to fasten a controller to it. (Continued under the cut.)

Impressions after the first days of use

From the obvious advantages:

• Well, this whole thing looks very impressive;
• Small size;
• Simple SDK. The documentation is normal, although it could tell you what the pitfalls of this version are. Examples are clear. It's pretty simple, we hook Listener to an object of class LeapController and everyone, we enjoy information about the hand, fingers, objects in the hand and gestures (appeared in the latest version);
• The data is really accurate (you can draw on graph paper with at least one finger).

')

Of the obvious cons:

• The device itself is very light, and the wire is hard, so it is very easy to move it;
• It is not convenient to use behind a laptop - the device is located at a small distance (vertically) from the monitor and to the lower parts of the screen is very difficult to get and no calibration has saved me .;
• When the device is heated (and if it is on a laptop, it comes very quickly) it starts to slow down and interfere with work (the mouse starts to twitch, the desktops switch);
• If you are close to a device, for example, if a head appears in its area of ​​view, Leap Motion often freezes;
• If the hand is turned vertically, so that the fingers are one above the other, the device does not understand this and thinks that there is only 1 finger;
• Bad recognizes the thumb;
• Does not recognize when the fingers are near, starts to simply blunt, recognizing one finger, then the other;
• The data is not filtered, because there is a “rattle” - this problem is on the shoulders of the developer.

Findings:

In general, the device is very interesting. But the creators of the work still car. It can’t replace the mouse - that’s for sure (even if they fix all the bugs), since it’s not convenient to hold hands on weight. Besides, it seems to me that they will correct some bugs and will not be able to because of the principle of work.

If you insert a normal battery and Bluetooth into the controller, then it can be very convenient to use it for remote control of a computer (TV), presentation, and similar kind of activity.

Although there are plenty of minuses, it’s very nice that such devices of this kind are beginning to appear not only on video but also in life!