This video is shows a new workflow tool I've developed for a recent project. In reality we used a real green screen setup and captured actual people. I don't have the footage from the project so Mr Lego man has stepped in to help explain.
In the project, we were creating animations of people at a party using exactly 4 frames for each (I've used 8 in this video). The footage coming from the camera was an image sequence with all of the people featured in a long stream of image files. The challenge was to batch process the footage so we end up with a series of 4 frame animations which we would then be looping. Now, I'm sure there are loads of ways to do this but I was unaware of any at the time and I spotted the opportunity to use the new R17 take system.
Essentially the script will look at the length of the image sequence, look at a parameter set by the user to say how long the animations should be and then create as many takes as it can based on the numbers. Each take has different parameters in the bitmap shader for start and end frames. Then, using the save tokens you can render out all of the animation files to movie files named in a sensible way. Have a look in the files for more information.
This is currently a very specific tool for a very specific job but you can adapt the parameters in the python script to make it do anything you'd like.
If you do find a good use for this then I'd love to hear what you've done.
Get the PIP - Batch take maker here : pixelsinprogress.com/downloads/