HOME | DD

Description
For as clear a pose as the original is, the AI did tend to often get the arm positioning wrong on this. There were also sometimes problems with the whole "being stuck" concept (like the AI producing images with doorways that can clearly be fit through), but that's to be expected since it's not a very common idea.

Again for these I had to mostly keep the denoising low - AKA the results constrained close to the original image.

The base image is cropped from this image: www.deviantart.com/zel1302/art… (can't share, because it's not something I have permission to share!)

Disclaimer: img2img is an iterative process, usually involving feeding output images back into the model to come closer to the ultimate image. While the starting image and generation data are provided, you shouldn't expect the base image to directly yield the final images seen here.

Number 1
Prompt: Morbidly obese woman too fat to fit through door, stuck in doorway, massive belly, green bra, ripped jeans, blonde hair, beautiful face
Seed: 2645427891
Other: Steps: 20, Sampler: Euler a, CFG scale: 8, Denoising strength: 0.2, Mask blur: 4, Size: 512x512
Model: Stable Diffusion 1.5

Number 2
Prompt: Morbidly obese woman too fat to fit through door, stuck in doorway, massive belly, green bra, ripped jeans, blonde hair, beautiful face, art by artgerm
Seed: 3584861944
Other: Steps: 20, Sampler: Euler a, CFG scale: 8, Denoising strength: 0.2, Mask blur: 4, Size: 512x512
Model: Stable Diffusion 1.5

Number 3
This one had a higher freedom than the rest - a denoising strength of 0.6, which is about as far as I'd reasonably let things go before the AI mostly starts doing its own thing rather than pay attention to the input image. Since this maintained most of the essence of the original input, I'm happy to include it here, rather than in the set of bonus images that were nice, but too far from the actual starting image to include here (which I'll upload at the end of this).

Prompt: Morbidly obese woman too fat to fit through door, stuck in doorway, massive belly, green bra, ripped jeans, blonde hair, beautiful face, hyper detailed, character concept, intricate, elegant, digital painting, artstation, smooth, sharp focus. A digital concept art by artgerm and greg rutkowski and alphonse mucha.
Negative prompt: deformed
Seed: 3584861944
Other: Steps: 20, Sampler: Euler a, CFG scale: 8, Denoising strength: 0.6, Mask blur: 4, Size: 512x512
Model: Stable Diffusion 1.5

Description
For as clear a pose as the original is, the AI did tend to often get the arm positioning wrong on this. There were also sometimes problems with the whole "being stuck" concept (like the AI producing images with doorways that can clearly be fit through), but that's to be expected since it's not a very common idea.

Again for these I had to mostly keep the denoising low - AKA the results constrained close to the original image.

The base image is cropped from this image: www.deviantart.com/zel1302/art… (can't share, because it's not something I have permission to share!)

Disclaimer: img2img is an iterative process, usually involving feeding output images back into the model to come closer to the ultimate image. While the starting image and generation data are provided, you shouldn't expect the base image to directly yield the final images seen here.

Number 1
Prompt: Morbidly obese woman too fat to fit through door, stuck in doorway, massive belly, green bra, ripped jeans, blonde hair, beautiful face
Seed: 2645427891
Other: Steps: 20, Sampler: Euler a, CFG scale: 8, Denoising strength: 0.2, Mask blur: 4, Size: 512x512
Model: Stable Diffusion 1.5

Number 2
Prompt: Morbidly obese woman too fat to fit through door, stuck in doorway, massive belly, green bra, ripped jeans, blonde hair, beautiful face, art by artgerm
Seed: 3584861944
Other: Steps: 20, Sampler: Euler a, CFG scale: 8, Denoising strength: 0.2, Mask blur: 4, Size: 512x512
Model: Stable Diffusion 1.5

Number 3
This one had a higher freedom than the rest - a denoising strength of 0.6, which is about as far as I'd reasonably let things go before the AI mostly starts doing its own thing rather than pay attention to the input image. Since this maintained most of the essence of the original input, I'm happy to include it here, rather than in the set of bonus images that were nice, but too far from the actual starting image to include here (which I'll upload at the end of this).

Prompt: Morbidly obese woman too fat to fit through door, stuck in doorway, massive belly, green bra, ripped jeans, blonde hair, beautiful face, hyper detailed, character concept, intricate, elegant, digital painting, artstation, smooth, sharp focus. A digital concept art by artgerm and greg rutkowski and alphonse mucha.
Negative prompt: deformed
Seed: 3584861944
Other: Steps: 20, Sampler: Euler a, CFG scale: 8, Denoising strength: 0.6, Mask blur: 4, Size: 512x512
Model: Stable Diffusion 1.5