Spyke

Replies

aigen

Comment on

Images with more than one person

Reply in thread

Rendering at larger sizes can also make it more likely SD will generate multiple subjects. You can use the Regional Prompter or Latent Couple extensions for stable-diffusion-webui to control the subjects separately.

You can also generate subjects separately using txt2img, paste them into a single scene, then do a pass with img2img to make the image look seamless. That is about the hardest approach, though. ControlNet with OpenPose and region prompter are my go-to solution.

nsfwai

Comment on

Reversed! (body horror)

Oh, here's the generation info:

a beautiful naked woman stands with her (back to the camera:1.4), shoulder blades, back of head, rear view, looking away, hand on other woman's butt ADDCOL a beautiful naked woman stands facing the camera, beautiful face, chest, breasts, nipples, clavicle ADDROW a beautiful naked woman stands facing the camera, navel, abdomen, pussy, hipbones ADDCOL a beautiful naked woman stands with her back to the camera, rear view, butt, ass, other woman's hand on her ass Negative prompt: blurry, airbrushed, low quality, black and white, painting, side view, twisted, pubic hair, extra arms ADDCOMM breasts, chest, clavicle, looking over shoulder, distorted face ADDCOL back, shoulder blades, looking over shoulder, spine ADDROW butt, back, spine, tailbone, hand, navel jewelry ADDCOL navel, hipbones, front, extra fingers Steps: 40, Sampler: Euler a, CFG scale: 7, Seed: 558380791, Size: 1024x1024, Model hash: 9584e2c050, Model: cyberrealistic_v32, Denoising strength: 0.1, ENSD: 31337, RP Active: True, RP Divide mode: Matrix, RP Matrix submode: Horizontal, RP Mask submode: Mask, RP Prompt submode: Prompt, RP Calc Mode: Attention, RP Ratios: "1,1,1;1,1,1", RP Base Ratios: 0.2, RP Use Base: False, RP Use Common: False, RP Use Ncommon: True, RP Change AND: False, RP LoRA Neg Te Ratios: 0, RP LoRA Neg U Ratios: 0, RP threshold: 0.4, Hires upscale: 2, Hires steps: 10, Hires upscaler: ESRGAN_4x, Version: v1.4.1-1-g81d08710

aikink

Comment on

Cyborg woman with SDXL

Reply in thread

I'd heard it was harder, but that may be with the base model. It's certainly been much easier so far when using SDXL Unstable Diffusers ☛ YamerMIX ! It doesn't work with tools like ControlNet, though, so it's limited to prompting and img2img. It might work for some limited use cases in Deforum that don't require ControlNet, like zooming with high strength. It's hard to get good videos in Deforum without zooming, though. (I'm trying to make videos with Deforum now.)

You reached the end