An experimental videoconferencing system is hoping to give users a shared space to interact. MM-Space, designed by researchers at Japanese telecom NTT, takes images of participants and edits out the backgrounds, then projects the images on transparent screens in a "recreation room." The result is a tripod-mounted face that appears (roughly) to be in the same room as the other participants. It's even able to turn to mimic the user's head motion, letting people know who she's looking at.

The premise of this actually seems excellent and would let videoconferencing more closely mimic its real-life counterpart. By removing the background, it reduces the sense of distance, and adding movements and eye contact could create more immersive telepresence. However, we're curious about its practical use. Will conference callers have to arrange disembodied face-screens around a room each time they want to talk? Will special videoconferencing rooms be arrayed with blank rotating panels? Even with all these issues, though, it would certainly make conference calls much more entertaining.