This evolution of Jeff Han's multitouch surface doesn't require touching, allowing you to manipulate objects Jedi-style thanks to infrared camera tracking overhead. This makes possible the use of conventional software, like a web browser, while keeping the advantages of direct touch manipulation. We talked about the technology and its advantages over previous multitouch displays with Florian Echtler, the researcher behind the project at the Technical University in Munich, Germany:
Jesús Díaz: Florian, this is an interesting evolution on the current multitouch solutions. What are the advantages of your method versus things like Han's displays or Microsoft Surface?
Florian Echtler: Well, the main difference is that you can control the cursor without touching the surface, so you can have the hover state which the usual mouse cursor has. For a "click" event, you have to tap the surface with any finger.
As a result of this, it's easier to use conventional software, like a web browser because a) you can use mouseover, which you can't with other touchscreens, and b) you can hit small targets (like a web link) more accurate, as you have the visual feedback from the cursor, showing you where you're going to click.
Second advantage: you can tell which hand a finger belongs to, and you can distinguish between fingers. You could, e.g., use the index finger for left click and the middle finger for right click—again, useful for conventional software emulation. But of course, you could just assign different actions to the fingers—think of a RTS game. We're not quite there yet, but on the other hand, it's still a prototype.
JD: I guess it requires less cleaning too, since you don't have to touch continuously or at all... How portable is this new method? How does it work?
FE: Unfortunately, it's not very portable right now, as it requires an additional overhead light source (also infrared). I assume you are familiar with FTIR (from Jeff Han)—this system is basically an extension, also taking the shadows from the overhead light into account.
JD: So it's not totally touchless, it requires touch to select and move things? Or can you move things without touching at all?
FE: While the video doesn't show this, it's possible nevertheless. Holding your hand still for a moment could be used for triggering an event.
JD: Why is there an offset in the video?
FE: In the first part, the offset is because we're displaying the raw, uncalibrated video data from the camera. In the second part, the data is calibrated and the cursors are deliberately shifted away from the hand to avoid occlusion.