Is it easier to chroma key directly on a switcher connected to a camera via HD-SDI or is it better to save that video in 50-mbit 4:2:2 and key it on After Effects?
I'd figure the video signal the switcher is being fed is pure, without compression, only chroma subsampling, and thus it's probably easier on the algorithm. Does this reasoning make sense?
However, I wonder if on AE you can do much more with time and other effects, specially if lighting of the chroma background is proper.
+ Reply to Thread
Results 1 to 5 of 5
99% of the time doing it later will be better. Chances are you don't have a perfect setup in terms of lighting. You can take your time and do other manipulations to improve the key, multiple mattes , fine tuning if you do it later
It depends on your specific camera and switcher, but a single HD-SDI should be 10bit 4:2:2 ; 50Mb 4:2:2 implies 8bit such has XDCAM-HD (but the 10bit isn't actually that useful for keying, the advantage is minimal over 8bit) . But not all cameras output true 10bit over SDI - some are "padded" 8bit in 10bit
But the one scenario where less compression helps tremendously is if your scene has lots of motion, or too taxing for the onboard compression. e.g. say some breakdancing vs. talking heads / static interview shot. But in that first scenario (or anything that is too much for internal compression) , I would use an external recorder, or rent / borrow one if you don't have one
But unless this is a brand new unreleased camera, probably someone has already done these sorts of camera testing - just search various camera forums, especially the ones for your model and manufacturer