Open-source face_lora_sd15 model - Focus on high-quality face image generation, free and easy to deploy

Face Lora Sd15

Developed by phil329

A fine-tuned version based on the Stable Diffusion architecture, utilizing Low-Rank Adaptation (LoRA) technology, focusing on high-quality face image generation.

Image Generation Open Source License:MIT #High-quality face generation #LoRA fine-tuning #Stable Diffusion optimization

Downloads 51

Release Time : 6/5/2024

Model Overview

This model is a fine-tuned version based on the Stable Diffusion architecture, optimized with LoRA technology, primarily used for generating high-quality face images.

Model Features

High-quality face generation

Trained on CelebA-HQ and FFHQ datasets, capable of generating high-quality face images.

Low-Rank Adaptation (LoRA) technology

Utilizes LoRA technology for model fine-tuning to enhance generation quality.

Supports negative prompts

Supports negative prompts to avoid generating low-quality or unrealistic images.

Model Capabilities

Text-to-image generation

High-quality face synthesis

Supports negative prompts

Use Cases

Artistic creation

Portrait generation

Generates high-quality portraits suitable for artistic creation and design.

Generated face images exhibit high realism and detail.

Virtual avatar design

Virtual character design

Used for designing character appearances in games or virtual reality.

Generates diverse face images to meet various character needs.

🚀 pytorch_lora_weights.safetensors

This is a fine - tuned Stable Diffusion model using LoRA, trained on high - quality face datasets for detailed facial synthesis.

🚀 Quick Start

You can quickly start using this model with the following code example:

import torch
from diffusers import StableDiffusionPipeline,UNet2DConditionModel

pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5").to("cuda")

pipeline.load_lora_weights("phil329/face_lora_sd15", weight_name="pytorch_lora_weights.safetensors")

NEGATIVE_PROMPT = "worst quality, low quality, bad anatomy, watermark, text, blurry, cartoon, unreal"
text = 'A young woman with smile, wearing a purple hat.'

lora_image = pipeline(text,negative_prompt=NEGATIVE_PROMPT).images[0]

display(lora_image)

✨ Features

This model is a fine - tuned version of the Stable Diffusion architecture, leveraging the Low - Rank Adaptation (LoRA) technique. It has been trained using the CelebA - HQ and FFHQ datasets, which are well - known for their high - quality images of human faces.

Training Details:

Property	Details
Base Model	Stable Diffusion
Adaptation Technique	Low - Rank Adaptation (LoRA)
Datasets	CelebA - HQ (30,000 images), FFHQ (70,000 images)
Resolution	512*512 fine - tuning for detailed facial synthesis

💻 Usage Examples

Basic Usage

import torch
from diffusers import StableDiffusionPipeline,UNet2DConditionModel

pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5").to("cuda")

pipeline.load_lora_weights("phil329/face_lora_sd15", weight_name="pytorch_lora_weights.safetensors")

NEGATIVE_PROMPT = "worst quality, low quality, bad anatomy, watermark, text, blurry, cartoon, unreal"
text = 'A young woman with smile, wearing a purple hat.'

lora_image = pipeline(text,negative_prompt=NEGATIVE_PROMPT).images[0]

display(lora_image)

Advanced Usage

We use four prompts as follows:

'A young woman with smile, wearing a purple hat.'
'A middle - aged man,beard ,attractive'
'A girl with long blonde hair'
'An young man with curry hair'

The negative prompt are the same as the example codes. All the results are randomly generated and not cherry - picked.

If the generation effect is not good, try adding a negative prompt, or try different prompts and seeds.

Result

📦 Installation

The weights for this model are available in Safetensors format. Download them in the Files & versions tab.

📄 License

This model is licensed under the MIT 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