I'm having a really tough time keying a DV source. I mean, the key is swell, but I'm getting serious stairstepping on the face and brighter areas. Using Boris, I can choke that away, but obviously, I'm cutting away a significant part of the image/spokesperson.
This was shot uisng a JVC GY DV-500.
My question is this: If I shot on a BetaSP camera and dubbed the resulting video to my Sony DSR45 (DVCam), am I back to where I started? Or will using component out (from a UVW1800) to the 45 give me better/more keyable results?
I've done pretty serious searches on keying for DV, but the only "tricks" that I came away with are to make sure the bkg is evenly lit and to use an amber backlight. Are there any other concepts that would make keying a person more doable in DV, or are the limitations of this format just too extreme.
The final result will have to be full NTSC, not web delivery.
Thanks for any and all advice, ideas and comments.
This was shot uisng a JVC GY DV-500.
My question is this: If I shot on a BetaSP camera and dubbed the resulting video to my Sony DSR45 (DVCam), am I back to where I started? Or will using component out (from a UVW1800) to the 45 give me better/more keyable results?
I've done pretty serious searches on keying for DV, but the only "tricks" that I came away with are to make sure the bkg is evenly lit and to use an amber backlight. Are there any other concepts that would make keying a person more doable in DV, or are the limitations of this format just too extreme.
The final result will have to be full NTSC, not web delivery.
Thanks for any and all advice, ideas and comments.