Video processing and recombining with missing faceframes?

[exnex]

Hello,

Not sure if it's possible to fix, or create a workaround..

I'm using vid2vid w/ lip retargeting.
My videos are nearly identical, I used live action vid to create a vid in anime style, but mouth detail was lost.
I'm trying to use the original live action as driving images, and the anime version as src vid, to create a lip sync effect

the problem is, the frames line up perfectly, but the face is sometimes not visible in the videos. This causes an error in retargeting node. I don't want to have to split both videos around missing face to keep frames aligned, process, then recombine.

Is there a way to ignore this error and force the process? or skip any pair of frames missing a face? track index of face detected frames and only process those, then reinsert into image batch at tracked index? any ideas would be helpful, i want to add this to a larger automated workflow where frame count needs to be dynamic. thank you