Ash at CIID

Ashwin Rajan's blog while at the Copenhagen Institute of Interaction Design.

Posts Tagged ‘computer vision

Toy View Workshop – Computer Vision Project 1

leave a comment »

Continuing from describing the intent of the fascinating Toy View workshop in my last post, here are some images from the first experimental project we built Рa  collaborative dance game called Face-Off. The game was developed using motion tracking in Adobe Flash. It is played by two dancers who are prompted to dance or move based on visual feedback (while dance music plays in the background). Only the dancer who is prompted visually must move, while the other must stay absolutely still. The team of two dancers wins if they are each able to dance at the right time and keep the music going until the end of the song.

Initial sketches of the Face Off game

Initial sketches of the Face Off game

Simple storyboards to work out the specifics of gameplay.

Simple storyboards to work out the specifics of gameplay.

Working out details in motion tracking on the webcam of a laptop.

Working out details of motion tracking on the webcam.

Screenshot of the game, which is projected on a vertical surface such as wall. A green box appears around the player whose turn it is to move.

Screenshot of the game, which is projected on a vertical surface such as wall. A green box appears around the player whose turn it is to move.

Toy View Workshop – Computer Vision Project 2

leave a comment »

In my last post I described the intent of the Toy View workshop held at CIID in December ’08, taught by Yaniv Stiener . Here are some of the projects we developed in the course of the workshop.

Reactable Game
Here is a video on the fascinating open-source Reactable technology – a collaborative music instrument with a multi-touch tangible surface-based interface – developed in Barcelona, Spain.

reacTIVision is an open source, cross-platform computer vision framework for the fast and robust tracking of fiducial markers attached onto physical objects, as well as for multi-touch finger tracking. The reacTIVision Engine and sample code, as well as Feducials (the markers used for tracking that can be attached to physical objects) can all be downloaded for free here.

//mtg.upf.es/reactable/?software)

How the reacTIVision technology works (courtesy http://mtg.upf.es/reactable/?software)

//mtg.upf.es/reactable/?software)

Annotated markers called 'Fiducials' can be pasted onto physical objects to faciliate tracking of object location and movements. (Courtesy http://mtg.upf.es/reactable/?software)

We built our own reacTIVision set up to develop a simple game during the course of this short workshop.

The reactable set-up

Experimenting with the reactable set-up.

We developed  a simple game using images of CIID students and faculty including Yaniv. The game mixed the heads, torsos and legs of different people depending on the how the feducials were placed and turned. A lot of fun and laughs all around!

Reactable game of mixed body parts.

Screenshot of reactable game that mixes body parts from different pictures.

Written by Ashwin Rajan

February 1, 2009 at 10:11 pm