I am working on a stadium shot.

I have a series of characters which each have the same skeleton/rig. I also have a series of BVH mocap files which I have converted to .bclip files. Each mocap will represent a state in the crowd simulator (i.e. sit idle, stand idle, waving, cheering, booing etc…) I have written a small python script to “retarget” the mocap on to the rigs. This is working ok.

Let's say I have 10 character and 10 mocaps. Using the shelf tools I can load each character 10 times and set my retarget code to the appropriate mocap and bake out each agent state.

With the example of 10x10 I would need to manually do this 100 times. But I actually have more than 10 characters and more than 10 mocaps so I am facing a massive amount of manual setup to convert these rigs into agent states.

Is there any way to automate this process?