{"id":1342,"date":"2018-11-28T22:16:37","date_gmt":"2018-11-29T03:16:37","guid":{"rendered":"http:\/\/www.joshuarosenstock.com\/teaching\/imgd3200-b18\/?p=1342"},"modified":"2018-11-28T22:16:37","modified_gmt":"2018-11-29T03:16:37","slug":"nick-sorensen-computer-vision-gestures","status":"publish","type":"post","link":"https:\/\/www.joshuarosenstock.com\/teaching\/imgd3200-b18\/nasorensen\/nick-sorensen-computer-vision-gestures\/","title":{"rendered":"Nick Sorensen – Computer Vision Gestures"},"content":{"rendered":"

For this project I took inspiration from the meme assignment. I used faceOSC to track the height of my mouth to determine when my mouth was open or not. This information was used to either play or stop a sound file using the playlist object. I had to experiment a bit to get the constant updating from faceOSC to only play the sound once. I used two debouncers that had to reset each other. FaceOSC is not perfect and sometimes does not detect your face correctly which leads to your mouth not being scaled right.<\/p>\n

 <\/p>\n

Below is the example template for faceOSC that was given. This just sends out the mouth height.<\/p>\n

\"\"<\/p>\n

Here is the max patch that reads the mouth height, compares it, and denounces it. When your mouth is detected as open it randomly selects one of the sound files.<\/p>\n

\"\"<\/p>\n

 <\/p>\n

Here is a demo of the patch in action.<\/p>\n