performance in full screen #14

Closed
heaversm opened this Issue Nov 10, 2012 · 2 comments

Comments

Projects
None yet
2 participants

Hi - I'm trying to use facesubstitution in full screen mode and I've noticed that if I trigger fullscreen (or even if I change the camera grabber dimensions and app dimensions, face substitution has a really hard time finding the face. Any idea why? Or if there's a way to still keep the capture dimensions small, but boost the display to full screen?

Collaborator

kylemcdonald commented Dec 10, 2012

if the camera is running at a higher resolution, then the face tracker will work harder to give you a more accurate placement of the tracked face. if you're not worried about getting more accuracy, then you can use the setRescale() function to something smaller to avoid the extra computation.

Thanks, I'll give it a shot.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment