Zukki_style Open-source Model - Fine-tuned based on Stable Diffusion to mimic the animation style of *The Life of Zucchini*

Zukki Style

Developed by tuwonga

A model fine-tuned based on Stable Diffusion v1.5, imitating the style of the stop-motion film 'My Life as a Courgette'

Image Generation Open Source License:Openrail #Stop-motion animation style #Character-specific generation #Img2img optimization

Downloads 93

Release Time : 11/15/2022

Model Overview

This is an experimental image generation model specifically designed to produce character images resembling the style of the stop-motion film 'My Life as a Courgette', performing exceptionally well in img2img mode

Model Features

Unique animation style

Specifically mimics the visual style of the stop-motion film 'My Life as a Courgette'

Img2img optimization

Performs better in image-to-image conversion mode than in pure text generation mode

Character generation specialization

Particularly adept at generating character images rather than scenes or landscapes

Model Capabilities

Text-to-image generation

Image style transfer

Character image generation

Use Cases

Creative arts

Animation-style character creation

Generate character images with a unique stop-motion animation style

Can produce character images resembling the style of 'My Life as a Courgette'

Image style transfer

Convert ordinary photos into stop-motion animation style

Achieve better conversion effects through img2img mode

🚀 zukki_style

A fine - tuned Stable Diffusion model based on v1.5, trained on screenshots from the stop - motion animation movie "Ma vie de courgette".

🚀 Quick Start

This is a fine - tuned Stable Diffusion model (based on v1.5) trained on screenshots from Ma vie de courgette stop motion animation movie. Use the token zukki_style in your prompts to use the style.

Download the ckpt file from the "files and versions" tab into the stable diffusion models folder of your chosen web - ui.

Actually, this is an experimental model because it mainly renders characters instead of scenes/landscapes. However, I've found the output in img2img more interesting than in txt2img. You can see the results in the second and third pictures (original/img2img/img2img). I think it would be better to check the "restore faces" option and play around with the denoising strength.

✨ Features

Character Rendering

With the Model: Prompt and settings used: [person] in zukki_style | Steps: 30, Sampler: Euler, CFG scale: 7.5
With img2img (First Example): Prompt and settings used: [person] in zukki_style | Steps: 30 - you can play around with settings
With img2img (Second Example): Prompt and settings used: [person] in zukki_style | Steps: 30 - you can play around with settings

Training Details

This model was trained with Dreambooth training by TheLastBen, using 32 images at 6400 steps with 25% of the text encoder.

📄 License

This model is open access and available to all, with a CreativeML OpenRAIL - M license further specifying rights and usage.

The CreativeML OpenRAIL License specifies:

You can't use the model to deliberately produce nor share illegal or harmful outputs or content.
The authors claim no rights on the outputs you generate. You are free to use them and are accountable for their use, which must not go against the provisions set in the license.
You may re - distribute the weights and use the model commercially and/or as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the CreativeML OpenRAIL - M to all your users (please read the license entirely and carefully).

Please read the full license here

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご