Got questions or tips about the Conversion process? This is the place to discuss them.
Read the FAQs and search the forum before posting a new topic.
This forum is for discussing tips and understanding the process involved with Converting faces from your trained model.
If you are having issues with the Convert process not working as you would expect, then you should post in the Convert Support forum.
Please mark any answers that fixed your problems so others can find the solutions.
- Posts: 1
- Joined: Mon Jan 11, 2021 2:09 pm
I was just wondering if there was any way of resizing an alignments file, which was extracted at 4k, for use with a convert to 1080p? As far as I can tell the alignments generated from a 4k video should be more accurate than those at 1080, but converting a 4k video turns out to be a real pain, at least for my system. I don't know exactly how the alignments file stores the data, but if it stores coordinates for each point, wouldn't it be as simple as dividing each value by 2?
Thanks for reading.
- Posts: 1503
- Joined: Fri Jul 12, 2019 12:53 am
- Has thanked: 54 times
- Been thanked: 288 times
I have never tried this, so there may well be caveats, but in theory that should work ok.
For instance, I'm fairly sure masks wouldn't work (as we store the transformation matrix for applying the mask back to the original frame).
Ultimately, you'd need to script this as it's unlikely to be something I would add to the code.