Faceswap Forum

There was an error reading from the Nvidia Machine Learning Library. Either you do not have an Nvidia GPU (in which case this warning can be ignored) or the most likely cause is incorrectly installed drivers. If this is the case, Please remove and reinstall your Nvidia drivers before reporting.Orig...

Yeah. Not how it works at the moment.

If you just want to get a face shape, then use the "cv2-dnn" aligner and generate a new face. It's not a good detector, but it will always give you a face shape. You can the edit these points.

katjagr wrote: ↑Thu Mar 04, 2021 11:59 am
Hi ... where did the alignments merge function within the tools/alignments tab go?
Is there any replacement function?

Please search first. This is already answered (in this thread).

For future reference: You MUST provide this file if seeking assistance. Please verify you are running the latest version of faceswap before reporting However, can you please provide me with the image: C:\Users\35484\Pictures\Ný mappa (2)\ezgif-3-2a26225ecbb8_001460_0.png (zipped, not directly attach...

Most likely you won't get any responses to this, as $500 is not a lot for the massive amount of work required for creating a Deepfake. Also a little more detail would not go amiss. However, I will leave this posted any way.

Which desktop environment are you using?

You can access faceswap from cli:

Code: Select all

conda acrivate faceswap
cd faceswap
python faceswap.py -h

Yes, there are copy "alignments" buttons on the left hand toolbar

Unfortunately none of the Devs run macOS. There are reports of some people getting it to work, and others having issues. One thing you do need is XQuartz installed for macOS. Otherwise, follow these manual install instructions (They are written for WIndows, but the process is the same for macOS): ht...

Downloading resources as they are needed is kinda standard for a lot of software, especially in the ML space. Why do we do it this way? To avoid the install package being tens of gigs in size, and so I don't have to compile, test and upload a new install package on every minor update. I do this for ...

Ok, I lied.

I did find this bug, and it should now be squashed.

That means there are missing Extended masks on your B side, but your A side is ok. It was meant to inform you that the A side has no issue, but I realize how that might be confusing, so I will remove the output line if there are no problems with the mask.

Yep. That's the idea. I want to remove the complication of an alignments file serving 2 purposes, and this is a step towards that.

https://github.com/alievk/avatarify

Ok. That is weird. For one thing, extended is auto-generated.

Can you provide the crash report please.

You can try forcing the regeneration of all masks by specifying process all.

Otherwise, I'm not sure what the issue is here, as I have no outstanding reports of issues in this area.

Not sure I really understand the question.

Alignments files are no longer required for training. Once your faceset is updated, the alignment data is loaded from the png headers

That option is no longer in the sort tool, as it is no longer required.

If you need to force sorting on CPU, then check all of the GPUs under the "Exclude GPU" option

Editing masks is really entirely up to you. Do what works, it will basically impact the final quality of the result, but it is quite tedious. An improved method is on my extremely lengthy "to do" list, but is unlikely to be addressed any time soon. re: DFL, it was trained on "legacy&q...

This bug should now be squashed.

The one you originally defined in "train". The others are backup snapshot folders that you can delete if you don't need them.

Faceswap Forum

Search found 1188 matches

Re: [Guide]Linux Install Guide: Installer Method

Re: Copy Landmarks

Re: alignments merge function

Re: Critical error when starting training

Re: Creatotor needed.

Re: [Guide]Linux Install Guide: Installer Method

Re: Copy Landmarks

Re: Mac OS High Sierra installing problem

Re: Why is there no offline installer?

Re: Save As bug

Re: Mask problem

Re: Updating facesets during training now alignments are no longer required

Re: What Is this deepfake?

Re: Mask problem

Re: Mask problem

Re: Missing Alignment Merge Job

Re: [Guide] Extraction - A Workflow

Re: Manual mask editing

Re: Could not broadcast input array from shape

Re: Multiple models directory