One option would be to have an input replay alongside captured outputs (audio visual), at some fixed framerate. Capturing intermediates (scene graph etc.) would probably also be valuable, as that could help nail down why something is failing.

Or you could do it [as I recall the project being called] the scientist way. You still have the old code, so you could replay inputs against each and compare. Probably more realistic because uncompressed video would be a ridiculously huge dataset. This would be more resilient in the face of testing hardware and driver drift.

Historically game engines are the worst offenders when it comes to unit testing. I'm not sure if that's still the case - but that's why I erred on the side of integration tests.