Hello to all, i wonder does anyone have experience using openCV for pinch gesture like this .
http://www.debralauterbach.com/Gestu...FinalPaper.pdf
I wonder what is the working principles/idea behind the project for this URL above.
Printable View
Hello to all, i wonder does anyone have experience using openCV for pinch gesture like this .
http://www.debralauterbach.com/Gestu...FinalPaper.pdf
I wonder what is the working principles/idea behind the project for this URL above.
Looks like reverse blob recognition to me. Instead of recognizing motion blobs, it detects background blobs. I did something similar once, the artical seems pretty straightforeward, I'm not sure how much I could improve on explaining the concept.
Why detect the background blobs since i need to get the pinch gesture hole ?
Please share your idea.
Well the pinch gesture hole is a background blob, so you need to detect the background blobs. Then you just need to set it to ignore small blobs that would be false positives from things like shadows and stuff.