Faceswap Forum

Your model file is corrupted. More info:
search.php?keywords=bad+object+header+version+number

It is not an ideal setup, no. However it should work just fine, but will likely be a bit slower than a more modern setup, and you will be limited to lower resolution models.

However, if you are just looking to learn the process, then this should work ok for you.

I'm assuming you mean 'forehead'? You can do this by using one of the landmark based masks (which are automatically generated during extraction). See here (scroll down a bit) for a visual representation of the masks:
viewtopic.php?t=27#extracting

Please provide more info as per: app.php/rules#rule-4b

It's not a bug, it will be to do with how you generated the mask in its configuration options.

If you select "include hair" then a head mask will be generated, otherwise a face mask will be selected. The default is to generate a face mask.

Most likely a missing dependency....

Are you able to post the output from the following command:

conda activate faceswap
conda list

Also, what version of macOS are you running?

There's no reason why it couldn't be done, although I am unfamiliar with the specific code for the destination.

It would require some coding to convert the model though, and potentially some coding to import the model at the other end.

There probably won't be a lot in it, to be honest.... If it were me, I'd probably get the 3090, but you'll likely be ok either way

I'm not really sure what you are asking.... How can you evaluate how accurate a swap is? If so, you can't. There are no real world 'swapped face' examples for you to compare against. You can only compare true vs generated for A>A or B>B. Not for A>B

My previews seem to work fine in testing,

Can you post the output of "Generate" (rather than "Extract") from the GUI, so I can test your explicit set up?

Your best bet will be to use the last version of Faceswap that supported PlaidML for AMD GPUs:

https://github.com/deepfakes/faceswap/r ... tag/v2.2.0

Well, I said it would be trivial. Issue was with files with multiple underscores (_) in the name.

Please update. Should be fixed.

Jacksonjunge wrote: ↑Fri Apr 26, 2024 10:53 am

I trust that a solution is coming.

Not unless someone actually makes the effort to provide me with some data, it's not.

If anyone can provide me with a video, alignments file + steps to reproduce this issue, please do, as I have not been able to.