X-lab Post #2



It’s important to be able to read the player’s silhouette as accurately as possible to make the game fair. I need a way to easily extract the background.

Green Screen: I took some green screen paint and started to make a green screen. I the attempted to use Processing to control the cameras turning off auto gain and auto exposure. But with the initial calibration screen shot and the game screen shot, there maybe too much uncertainty.

IR Camera: I purchased a cheap IR webcam to test taking screen shots of silhouettes. If I can build a photobooth type box, I can control the lighting and take an IR photo of the player in the dark. This would cut back on the uncertainty, that the green screen method has.

Currently, I am building a mock photobooth to test lighting conditions.