Faceswap Forum

This is an "Out of Memory" error. If it worked before and doesn't now, then most likely something is taking precious VRAM away from your GPU (Google Chrome can be bad for this, also games etc.)

You should try following the steps listed in the error message output.

<- This means enlarging B's face, right? You don't need to enlarge it, no. But you are correct that the face will be enlarged (or shrunk) to the correct size to fit in the final frame. The size of the model output is uniform (be that 128px, 256px, 512px etc.) The size of any face in a frame could b...

I'm not sure I fully understand your problem. The first thing I would suggest, though, is following these steps as it solves 90% of issues: https://forum.faceswap.dev/app.php/faqpage#f1r1 If this doesn't fix this issue, then please can you provide me with more detail about your setup (OS, GPU etc) a...

Unfortunately (or fortunately?) I have never seen this issue before.

I would only be able to investigate with a copy of the video and alignments file. If this is something you can share, you are welcome to DM me with a link.

If I understand correctly, I think you are talking about applying a 256px face on high resolution close-ups? You are correct that close ups like these won't work as well as the face needs to be enlarged to fit into the frame. However, it is hard to say how much this will matter to you, and how bad i...

Don't worry about it (the short answer) The slightly longer answer: I'm not 100% sure where this comes from, but I think it comes from the preview image in the GUI. Most likely when the process tries to read a preview image when it hasn't yet been fully saved. The message itself is not something I c...

The reason is that the Conda environment needs to be activated. Just referencing the correct python library is not enough. When the Conda environment is activated, some other actions are also executed, which means that Faceswap can find the Cuda/cuDNN libraries stored within the environment. If you ...

I believe it works if you use Mixed Precision. The issue comes when switching from mixed precision to full precision. When you start a new model at full precision, Mixed Precision is activated (as we need to store which layers are compatible with Mixed Precision and this is the only way to do it). W...

Can you post a screengrab showing what you mean? This is a new issue on me.

Yes, unfortunately that advice on earlier version of TF is now out dated.

What I advise is:

load the StoJo preset
switch the Encoder to EfficientNetV2-S

This should work fine without needing to adjust encoder scaling (make sure you are creating a new model and not resuming an existing one)

Those message are standard, as we have to store mixed-precision compatible layers at start up. As it looks like you are using Apple Silicon, then that message is expected. I can't speak for speed, as I don't have an Apple computer, however, if you run training with VERBOSE logging, it should tell yo...

I love over analysing datasets, so good work, and keep it up :) Feel free to share your script here so others can try it out. I can also look at your blur code, as the blur detection we have in Faceswap already masks out the background, so if it works well, I can look to implement it as a sort optio...

Please do. The training 'plugin' is not really a plugin per-se.... It is just structured in a way that I can split out for new plugins etc. in future. None of the work has actually been done, because I wouldn't like to guess at what would/would not be useful, but the architecture is built in a manne...

Thanks for the prompt reply. I think I need to do some more reading first to make sure I’m not missing anything :) . There seems to be an implicit flow that I’m not quite seeing just yet. Not that there’s any actual faceswap reason to do something like this, but from a coding perspective, it doesn’...

I would add to this... We would very much welcome new models. We are not too fussy as long as it ultimately works ;) I would add the following for creating models: I would recommend looking at the Original (and maybe IAE) model layouts, as they will be the easiest to understand how the models get de...

1: After completing the steps of Clean the Alignments File, can I start training? Yes. that should be fine 2: Manually Fix Alignments It looks like I have rewritten the process of extracting alignment files again? So, when I execute Manually Fix Alignments, is the .fsa file I use just extracted usi...

So, assuming you install Faceswap using the old (deprecated) AMD PlaidML backend, then the models created will not be compatible with either the Nvidia or the CPU version. Unfortunately there is no way around that. The Nvidia/CPU versions used Tensorflow as the backend. The AMD version used PlaidML ...

Thanks for the update. I will still look into this when I have time, because it does not sound like it is working as expected. I've just been a bit short on time lately. There may be a bug with 'missing' in the mask tool not working correctly, so I will look into that too. Will update if/when I get ...

PM'd you a screen recording, google drive link. Though it seems to be lingering in the outbox. Just means I haven't picked it up yet. Most likely I will do tomorrow. Update: I notice that in some cases, magnifying the view changes the green square framing. (Not just rotating so it's upright, but th...

I've not seen this. Are you able to post a short video/screengrabs to show what you mean?

Faceswap Forum

Search found 2669 matches

Re: Models in different GPU/CPU

Re: Size of faces in B-side video source

Re: Stuck on loading screen

Re: During conversion, "No alignment found for xxxxx.png, skipping" only on some video files

Re: Size of faces in B-side video source

Re: What is "libpng error: Read Error"?

Re: GPU don't work when program aren't started in anaconda

Re: When using the preset stojo file in Phaze-A, the error is reported

Re: Preview only shows mask? Phaze-A, maskers: Bisect-Fp + Vgg-Clear + VggObstructed

Re: When using the preset stojo file in Phaze-A, the error is reported

Re: What's a reasonable speed? Also, error message

Re: My way of creating good training set

Re: Location of “forward pass” and/or Where to implement a Custom Training Loop

Re: Location of “forward pass” and/or Where to implement a Custom Training Loop

Re: Location of “forward pass” and/or Where to implement a Custom Training Loop

Re: [Guide] Extraction - A Workflow

Re: Models in different GPU/CPU

Re: Mask editor bug? Zoom in-out discrepancies

Re: Mask editor bug? Zoom in-out discrepancies

Re: Mask editor bug? Zoom in-out discrepancies