Although the Leap 3D sensor is still not available for general sale, the developer kits do seem to be producing some impressive results. It looks as if the Leap might be the great leap forward it promised to be.
Initially the I Programmer team was skeptical about the Leap because it seemed to claim far too much for too little effort and cash. A $70 device shouldn't be capable of 10 micron 3D resolution, which is what it claims to do.
We still don't know if the specifications are anywhere near the mark or if there is some limitation that isn't quite so obvious. However, as the first independent demos start to appear it becomes clear that the device does broadly what it claims to do, i.e. track accurately enough to follow the movements of individual figures on a hand.
If this seems like a small claim to you then you need to look at the inability of devices like the Kinect to do the same job and difficulties that people have to overcome to do the same job using known technologies. Of course, we don't know what technology the Leap actually uses.
In this example we have a quadrotor controlled by hand movements via a LabView program. The video shows the unboxing of the device and the building of a quadrotor controller in under 24 hours.
A second project, AirHarp, is perhaps a better example of precise finger detection. As the implementor, Adam Somers says:
Leap Motion is a highly precise and responsive motion tracking device, making it a perfect tool for expressive musical interactions.
You can see the air harp in action in the following video:
The code for AirHarp is on Github and its in C++.
So it looks as if Leap is a real user interface device that works well enough for these sorts of applications. What is important and game changing is the fact that the Leap works on the desktop and at close distances. With the Kinect you generally have to be a few feet away and it is best at whole body tracking. The Leap with its close and accurate operation might just provide an alternative to the long serving mouse or even provide a way to convert every screen to a touch screen.
Let's see what happens when the Leap is available to all.
Compression is less important these days for storage, but it is still a valuable way of saving bandwidth. Google has offered improved compression algorithms in the past and now we have Brotli, which c [ ... ]
This isn't big theory news like NP=P or anything similar, but it is a fascinating idea. About 80 years ago the remarkable and strange mathematician Paul Erdos formulated what seemed like a simple ques [ ... ]