[David May] “so that the subject remains stable, the footage remains in shot (no black bars) and no up-scaling is required.”
Just wanted to clarify that this is not really possible with any technique (unless you have acquired at higher resolution than you need to deliver).
If the acquisition raster is the same as the delivery raster, you’ll have to move the frame around to stabilize, which will reveal the layers beneath it (or the background color). To eliminate that, you’ll have to do something (scale or distort the original footage, or use motion tile, multiple layers with masks, etc.) to make up for pixels that just aren’t there in the camera footage.
Many of these techniques are about mitigating this issue, but it’s inherent in the process.
Walter Soyka
Principal & Designer at Keen Live
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
RenderBreak Blog – What I’m thinking when my workstation’s thinking
Creative Cow Forum Host: Live & Stage Events