IDisplay, Webcam Multitouch

June 26, 2009 by Eliot 32 Comments

[youtube http://www.youtube.com/watch?v=mlLY0zic7u0%5D

Embedded above is an interesting multitouch demo by [Lahiru]. The goal of the project was to find an easy way to retrofit current LCDs for multitouch. Instead of using infrared or capacitive recognition, it uses a standard webcam mounted overhead. To calibrate, you draw polygon around the desktop screen as the webcam sees it. The camera can identify the location of markers placed on the screen and their color. iDisplay can also recognize hands making the pinch motion and sends these as touch events via TUIO, so it works with existing touch software. It’s written in C++ using OpenCV for image processing with openFrameworks as the application framework.

[via NUI Group]

Pseudo 3D Chat

January 29, 2009 by Eliot 21 Comments

3d_video-conference

[Chris Harrison] and [Scott E. Hudson] have built a novel system for faking a 3D video chat session. Their implementation separates the image of the chat participant from the background. They then dynamically reposition the video based on the movement of the viewers head. Their using the OpenCV library to do facial recognition (just like the Laughing Man demo). The 3D effect is very similar to what you see in [Johnny Lee]’s Wiimote headtracking. A video of the pseudo 3D chat is embedded below.

Continue reading “Pseudo 3D Chat” →

Laughing Man In Processing

December 31, 2008 by Eliot 50 Comments

laughing_man

The Laughing Man is the antagonist from the anime series Ghost in the Shell: Stand Alone Complex. During each of his public appearances in the series he manages to hack all video feeds/cyborg eyes in the vicinity to obscure his face with the logo above.

[Ben Kurtz] had been watching the series recently and realized he could put together a similar effect using Processing. The interesting bit, and what makes this more fun than a simple demo, is that he’s using the OpenCV library. OpenCV is a open source computer vision library. [Ben] uses it to handle the facial recognition in Processing and then apply the image.

It’s only 100 lines and we wonder what other fun tricks could be employed. Here’s a Hack a Day skull you can swap in for the logo.

[thanks dakami]