(L10n_H34r7) (2024)

Image description: A girl in the foreground taking a selfie in dense jungle, with a roaring lion approaching in the background. She is wearing a yellow shirt, green shorts, and has a backpack on.

Full Generation Parameters:

Selfie, fisheye lens, photorealistic girl traveller screaming and running : in the background, angry lion, jungle

Negative prompt: fale, dull, boring, plastic, 3d render

Steps: 20, CFG scale: 2.5, Sampler: Euler, Seed: 648188728958536, Size: 832x1216, Model: flux1-dev-fp8, Version: ComfyUI, Model hash: 8e91b68084

  • azthec
    link
    fedilink
    English
    arrow-up
    1
    ·
    edit-2
    26 days ago

    Is it possible to generate this kind of funny image with your own face already? I know that there’s all the deepfake type stuff people do but I assume with an image generator you get better quality.

    This might be obvious but I’m very out of the loop on image generation

    • Even_Adder@lemmy.dbzer0.comOP
      link
      fedilink
      English
      arrow-up
      2
      ·
      26 days ago

      It’s possible with if you train an adapter on yourself or using ipadapter_FaceID. The model this was generated with is kind of new and doesn’t have many tools yet.