Faceswap Forum

Balmmaker

I saw that. The method though? Extract the faces, add them to the media folders... But then I need to add the faces how?
Tool —->sort - - >' missing alignments'?

It isn't exactly obvious how to make sure the new and altered data set is seen.

Balmmaker

What is the process for taking a model which has trained on [data person A] and [data person B] for say 400k iterations and then giving it new data for person A and person B? I am hoping this will improve the overall model, but what steps do I take to actually execute this?

Balmmaker

So a batch size of 40 is too low for dfaker? And for variety of images, is 8000 individual photos in both the swap and original sets, so 16000 photos total, you don't think there is enough variation in that? I can try and add to these photo sets or create a separate group of data to train on. What i...

Balmmaker

Here is a typical level of blurriness -- this is after 650k iterations at the settings described above. https://i.ibb.co/fn4FFk9/blurryfaceswap.png And here is a typical level of quality of source material before extracting the faces https://i.ibb.co/ZBf85Xr/faceswap-01.jpg I am not particularly att...

Balmmaker

I am seeking better settings to try for swapping photo model A with photo model B. I have over 10k high quality, good lighting images for both. Currently I am unable to get any sharp or at least not quite blurry and passable image swaps using 128x128 inputs, dfaker model, batch size of 40, 4000 face...

Balmmaker

Can someone post some pictures of swaps when they consider training complete and the swapped results actually turn out decent? I would like some sort of reference for when to stop. (Does the blurriness ever go away completely? And is there a way to tweak a model with 300k iterations to get it to sha...

Balmmaker

Ah, well the batch size is 16. The training images are all very sharp and from high quality pictures --so the data is good (i think!). I am now up to 450k iterations and it still looks like a 'blur' filter has been applied to the face, smudging it a bit. I attached a photo. Just wondering if I am on...

Balmmaker

I have been training for 60 hours now on at rtx2080 with 32gb of ram using realface 128k-->256k. Thus far, I have completed 281k iterations the faces appear to be heading in the right direction yet the swapped face is still pretty blurry and so more training is needed. My questions simply status bas...

Balmmaker

You are waaaay smarter than the average bear, boo boo! That seemed to work and the training is now running! TY TY!

(How you remembered that was awesome.) Thanks again.

Balmmaker

Fair enough. Here is the crash log. Ty for your patience as I really should have attached this log. 12/04/2021 02:35:16 MainProcess _training_0 generator __init__ DEBUG Initializing TrainingDataGenerator: (model_input_size: 128, model_output_shapes: [(256, 256, 3)], coverage_ratio: 0.875, color_orde...

Balmmaker

I am trying to kick off a training session but it crashes straight away. I am wondering if it has to do with tensorflow or python being the wrong version number? But I am kinda out of my league on this one. The extraction used VGG-obstructed and the global config for training also had VGG-obstructed...

Balmmaker

I think I figured it out. There seems to be a bit of a glitch I tested a few times. Since I was using photos for the sets, in the extraction process I had set N=2 because I wanted half the set done so I could test the results on the other half or use those images for training. However, this seems to...

Balmmaker

Nope. The previews in the thumbnails are representive of the very off wire frames :/ I redid the extraction too, did it again and had the same thing happen

Balmmaker

Here is some typical examples. The alignments are obviously not for the same image at all, for all of them.

Balmmaker

First time faceswapper so be kind. Using photos as data sets. 1 set worked. Issue is with the 2nd extraction. Extracted the faces successfully. Sorted them successfully (sorted by 'face', output to 'rename', 'keep' not checked, group by 'hist'). All the faces were sorted. Then I went into the sorted...

Faceswap Forum

Search found 15 matches

Re: Changing the training data after some learning

Changing the training data after some learning

Re: Best settings for photos to photos

Re: Best settings for photos to photos

Best settings for photos to photos

Faces picture when training is complete?

Re: Noob Training Progress Question

Noob Training Progress Question

Re: Critical Unexpected Crash: Custommaskwarning:

Re: Critical Unexpected Crash: Custommaskwarning:

Critical Unexpected Crash: Custommaskwarning:

Re: Alignments file out of snyc with images

Re: Alignments file out of snyc with images

Re: Alignments file out of snyc with images

Alignments file out of snyc with images