Best settings for photos to photos

[existenceproduct]

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 faces for each set - went through each and every image to ensure the wire frame face matches the photo and it positioned correctly.

Reasably good hardware with an Rtx2080 and 32GB RAM.

Even with 500k iterations, the preview is noticably blurry. The improvement from 100k to 500k iterations was small but present. Previous attempts ran more than 1 million iterations while worse blurry previews than 100k with realface (used the original model and 16 batch size, 64x64 input)

What I am wondering is, what is the best training settings to try for large photos sets with good lighting? About how many iterations ballpark should I wait till I see reasonable sharpness or abandon to different settings? Is there anything else I can do to sharpen things up? Faceswap can do photos well and not just videos, right?

 

[existenceproduct]

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 faces for each set - went through each and every image to ensure the wire frame face matches the photo and it positioned correctly.

Reasably good hardware with an Rtx2080 and 32GB RAM.

Even with 500k iterations, the preview is noticably blurry. The improvement from 100k to 500k iterations was small but present. Previous attempts ran more than 1 million iterations while worse blurry previews than 100k with realface (used the original model and 16 batch size, 64x64 input)

What I am wondering is, what is the best training settings to try for large photos sets with good lighting? About how many iterations ballpark should I wait till I see reasonable sharpness or abandon to different settings for Face swap? Is there anything else I can do to sharpen things up? Faceswap can do photos well and not just videos, right?

thanks in advance for any help

 

[Godfrey]

Honestly, I have no idea what you're talking about.

G

 

[f6cvalkyrie]

Honestly, I have no idea what you're talking about.

G

+1
Rafael

 

[buildbot]

Honestly, I have no idea what you're talking about.

G

+1
Rafael

I think they are talking about using machine learning (ML) to swap a model in one photo into another. It is image processing of a kind!

thanks in advance for any help

This is 99% a photography and related gear forum, you'd probably have better luck making a GitHub issue for whatever tool you are using or asking their discord server or whatever they use for support. Image processing does involve ML more often these days, but the vast majority of people here are going to be using it via photoshop or another GUI tool.

I wish I could help, but sadly my expertise is mostly in transformer models for natural language processing, and colorization on the image processing side of things, not faceswap / dfaker / style models as much.