Ahh, springtime. The sun is out, there’s a fresh breeze blowing, the scent of blooming flowers is in the air, birds are chirping, etc. etc. And the final FPGA projects from the spring 2013 semester keep coming in!
Today we’ve got a great one from Cornell University student Tian Gao, built using an Altera DE2 FPGA and some fancy hand-detection tech. Tian’s got the Altera FPGA capturing a video signal and detecting a player’s hand skin color. Tian says,
The system tracks their hand movements and looks for the user moving their hands forward and back by determining how tall the skin colored object is over time. The goal here is to prevent the falling tetris like blocks from falling by rapidly pushing and pulling on them.
[The project is] divided into three parts: video signal decoding, hand detection and game playing. The camera provides a standard NTSC signal for FPGA and the FPGA has a hardware decoder which generates a serial digital signal. Altera has provided a solution to decode it to RGB system and display it on VGA screen. For hand detection, I first detect skin in YCbCr scale, then I track the hand by an iteration algorithm. The game is the similar as the original tetris, with the difference in width and height of the screen. Also you can’t move or rotate the block, you can just eliminate it.
He calls his game creation Anti Tetris. Tian has documented the entire project here if you’d like to read up on the finer points of his efforts. This is some good work, and looks like it might actually be a bit of fun to play!
Leave a Reply
You must be logged in to post a comment.