dreambooth_diffusion_hokusai Open-source Image Generation Model - Create Paintings in the Style of Hokusai for Free

Dreambooth Diffusion Hokusai

Developed by keras-dreambooth

Text-to-image diffusion model based on Katsushika Hokusai's art style, fine-tuned using Dreambooth technology

Image Generation Open Source License:Openrail #Katsushika Hokusai art style #Text-to-image generation #High-resolution image generation

Downloads 18

Release Time : 3/18/2023

Model Overview

This model learns the artistic style concepts of Katsushika Hokusai through Dreambooth technology and can generate images with the artist's style based on text prompts.

Model Features

Specific art style generation

Specialized in learning and imitating Katsushika Hokusai's unique art style

Dreambooth fine-tuning

Fine-tuned for style concepts using Dreambooth technology

High-resolution output

Supports generating images at 512x512 and higher resolutions

Precise control

Allows exact control over generation effects through prompts and negative prompts

Model Capabilities

Text-to-image generation

Art style imitation

High-resolution image generation

Negative prompt control

Use Cases

Art creation

Hokusai-style landscape painting

Generate landscape paintings in Katsushika Hokusai's style

Examples show stylized depictions of fishing villages under cherry blossom forests at sunset

Surrealist art creation

Create surrealist works combining Hokusai's style

Examples show surreal depictions of lost valleys in mountains

Design assistance

Concept design

Provide stylized concept images for design projects

🚀 tf-keras

This is a fine - tuned model for text - to - image tasks, specifically learning the concept of the Hokusai artist.

🚀 Quick Start

This model, tf-keras, has been fine - tuned to capture the essence of the Hokusai artist's style. You can use it for text - to - image generation by providing prompts with the concept token "hks## style".

✨ Features

Fine - Tuned Concept: It has learned the unique style of the Hokusai artist from a specific dataset.
Text - to - Image Generation: Capable of generating images based on text prompts.

📚 Documentation

Model description

This model has been fine - tuned to learn the concept of Hokusai artist by ChienVM on this dataset

Intended uses & limitations

Concept token "hks## style". Example: "a photo of painting image in hks## style"

Some examples prompt

Sample 1 — Prompt: a painting image of a fishing village under a cherry blossom forest at sunset in hks## style, ultra realistic, 4k, 8k. Negative prompt: bad anatomy, soft blurry. Steps: 100, CFG scale: 15, Size: 512x12

Sample 2 — Prompt: a beautiful and highly detailed oil painting of a lost valley in the mountains in hks## style, running river, intricate details, 8k, sharp focus, hyper realism. Negative prompt: (bad anatomy), (blurry), grain. Steps: 100, CFG scale: 15, Size: 512x12

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

Property	Details
inner_optimizer.class_name	Custom>RMSprop
inner_optimizer.config.name	RMSprop
inner_optimizer.config.weight_decay	None
inner_optimizer.config.clipnorm	None
inner_optimizer.config.global_clipnorm	None
inner_optimizer.config.clipvalue	None
inner_optimizer.config.use_ema	False
inner_optimizer.config.ema_momentum	0.99
inner_optimizer.config.ema_overwrite_frequency	100
inner_optimizer.config.jit_compile	True
inner_optimizer.config.is_legacy_optimizer	False
inner_optimizer.config.learning_rate	0.0010000000474974513
inner_optimizer.config.rho	0.9
inner_optimizer.config.momentum	0.0
inner_optimizer.config.epsilon	1e - 07
inner_optimizer.config.centered	False
dynamic	True
initial_scale	32768.0
dynamic_growth_steps	2000
training_precision	mixed_float16

📄 License

This model is licensed under the creativeml-openrail-m license.

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