Project Question

Hello, I’m an art major and (majoring in 3-D modeling/animation and photography) I want to do a project using motion sensors. The project is to have a mirror or reflective surface (could be glass) and when people walk up to the reflective surface they will trigger the motion sensor device and instead of seeing their own reflection they will see the creatures and monsters I make in Mudbox. (3-D program that is very realistic. It’s basically sculpting clay but on a computer) The reflections and creatures would be a bust, so chest up. It would be amazing if the creatures could possibly open and close their mouths, open and close their eyes, or just even move when the person did. (don’t have to it would make the project more interesting) I have never done anything like this and certainty don’t know much about motion sensored devices or what the possible hook ups would be. I’m sure I could would use a projector and would have to use a computer for my images but I’m not certain the best route would be for a motion sensor device. It was brought to my attention through a fellow classmate the Xbox 360 Kinect is open sourced and I came across this video This is amazing and what I’m going for. Would using the Kinect be the best route for me or should I focus on webcam software or a real motion sensored device?

Thank you so much for your time and input. Much appreciated. I’m just trying to get information.


Hi Tawnee,

If you’d like to track people’s postures, the kinect is the way to go. This thread will be of interest:

If you’d just like to detect people’s presence or some rudimentary about their position, a webcam will suffice. There has been a lot of work done using modified play station eyecams, near infrared illumination , and the computer vision techniques of background subtraction and blob detection. The OpenCV example that comes with oF should get you started on background subtraction and blob detection.