4 ms·

how I wish it had a strength setting.. you can try blending the before and after with PIL but it will probably look even more disturbing strength = 0.5

by tlack 3y ago

how I wish it had a strength setting.. you can try blending the before and after with PIL but it will probably look even more disturbing strength = 0.5 horrific_image = Image.blend(image_with_normal_skin_and_teeth, kinda_awful_gfpgan_image, strength) let me know if you find something better

shipwright 3y ago
Fortunately, that is indeed a thing :) several post-processing tools have some sort of gfpgan_strength (searching as-is should yield results on a few repos) param being passed, also it's been a while since diffusion-based restoration methods are equipped to diminish stochasticity enough to produce great results, worth the time as well! Also noteworthy: https://openmodeldb.info/?t=faces+restoration+texture-generation+arch%3Adat+arch%3Ahat https://openmodeldb.info/?t=faces+restoration+texture-genera...
- tlack 3y ago
  In Auto1111, that just uses Image.blend. :) https://github.com/AUTOMATIC1111/stable-diffusion-webui/blob/cf2772fab0af5573da775e7437e6acdca424f26e/scripts/postprocessing_gfpgan.py#L28 https://github.com/AUTOMATIC1111/stable-diffusion-webui/blob...
  - shipwright 3y ago
    Hah, so high tech- Turns out the said repos used to do the same, then one of them later added CodeFormer under the hood, conflated the strength with affinity and didn't bother renaming (calling codeformer --gfpgan_strength looks very unhinged to me but oh well haha). It's definitely doable though! What I had in mind as a GAN equivalent (used to apply in WSI context while back) is akin to Rayleigh EigenDirections in terms of latent space navigation https://www.ecva.net/papers/eccv_2022/papers_ECCV/html/7277_ECCV_2022_paper.php https://www.ecva.net/papers/eccv_2022/papers_ECCV/html/7277_... There's some code out there, unsure if it's streamlined for this use case but can lend a hand if you decide to give it a shot :)

by tlack 3y ago

shipwright 3y ago

Fortunately, that is indeed a thing :) several post-processing tools have some sort of gfpgan_strength (searching as-is should yield results on a few repos) param being passed, also it's been a while since diffusion-based restoration methods are equipped to diminish stochasticity enough to produce great results, worth the time as well! Also noteworthy: https://openmodeldb.info/?t=faces+restoration+texture-generation+arch%3Adat+arch%3Ahat https://openmodeldb.info/?t=faces+restoration+texture-genera...

tlack 3y ago
In Auto1111, that just uses Image.blend. :) https://github.com/AUTOMATIC1111/stable-diffusion-webui/blob/cf2772fab0af5573da775e7437e6acdca424f26e/scripts/postprocessing_gfpgan.py#L28 https://github.com/AUTOMATIC1111/stable-diffusion-webui/blob...
- shipwright 3y ago
  Hah, so high tech- Turns out the said repos used to do the same, then one of them later added CodeFormer under the hood, conflated the strength with affinity and didn't bother renaming (calling codeformer --gfpgan_strength looks very unhinged to me but oh well haha). It's definitely doable though! What I had in mind as a GAN equivalent (used to apply in WSI context while back) is akin to Rayleigh EigenDirections in terms of latent space navigation https://www.ecva.net/papers/eccv_2022/papers_ECCV/html/7277_ECCV_2022_paper.php https://www.ecva.net/papers/eccv_2022/papers_ECCV/html/7277_... There's some code out there, unsure if it's streamlined for this use case but can lend a hand if you decide to give it a shot :)