Source image batches, and new filename tokens

[pope7]

My current issue:
Because this tool uses a single source image as the starting point, I find myself trying dozens of potential shots of a single person and then comparing the output. Currently, I have to change the input image one at a time, manually.

Potential solution:

1. Just like we have now with multiple video targets, allow for multiple input images.
2. Allow for ctrl/shift-clicking faces in the face pane to select multiples

Additionally, adjacent to the above:
I'd adore the ability to set my output filename with tokens for both the [source image filename] and [target video filename]. This would ease results comparison, as I could have a folder full of TARGET1-FILENAME1.mp4s and it'd be clear what was being tested from the filename.

Love the work, thank you!

[pope7]

This has finally beaten it, but before switching I was using a (not on github) fork of roop that allowed this multi-dimensional batching and filename structuring. I'd link the code, but I think it died with the death of Roop's Discord.