Controlnet Openpose Sdxl 1.0 Safetensors

🚀 SDXL-controlnet: OpenPose (v2)

These are controlnet weights trained on stabilityai/stable-diffusion-xl-base-1.0 with OpenPose (v2) conditioning, enabling text-to-image generation with specific pose control.

🚀 Quick Start

This is a copy of thibaud/controlnet-openpose-sdxl-1.0, allowing direct usage from diffusers using the safetensors version.

✨ Features

• Trained on stabilityai/stable-diffusion-xl-base-1.0 with OpenPose (v2) conditioning.
• Can generate images with specific pose control, like making Darth Vader dance.
• Can be used in 🧨 diffusers library.

📦 Installation

First, install all the libraries:

pip install -q controlnet_aux transformers accelerate
pip install -q git+https://github.com/huggingface/diffusers

💻 Usage Examples

Basic Usage

Now, we're ready to make Darth Vader dance:

from diffusers import AutoencoderKL, StableDiffusionXLControlNetPipeline, ControlNetModel, UniPCMultistepScheduler
import torch
from controlnet_aux import OpenposeDetector
from diffusers.utils import load_image

# Compute openpose conditioning image.
openpose = OpenposeDetector.from_pretrained("lllyasviel/ControlNet")

image = load_image(
    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/person.png"
)
openpose_image = openpose(image)

# Initialize ControlNet pipeline.
controlnet = ControlNetModel.from_pretrained("dimitribarbot/controlnet-openpose-sdxl-1.0-safetensors", torch_dtype=torch.float16)
pipe = StableDiffusionXLControlNetPipeline.from_pretrained(
    "stabilityai/stable-diffusion-xl-base-1.0", controlnet=controlnet, torch_dtype=torch.float16
)
pipe.enable_model_cpu_offload()

# Infer.
prompt = "Darth vader dancing in a desert, high quality"
negative_prompt = "low quality, bad quality"
images = pipe(
    prompt, 
    negative_prompt=negative_prompt,
    num_inference_steps=25,
    num_images_per_prompt=4,
    image=openpose_image.resize((1024, 1024)),
    generator=torch.manual_seed(97),
).images
images[0]

Example Images

Here are some example images:

• prompt: a ballerina, romantic sunset, 4k photo

Comfy Workflow

(Image is from ComfyUI, you can drag and drop in Comfy to use it as workflow)

Generated Examples

Here are some generated examples:

📚 Documentation

Training

Use of the training script by HF🤗 here.

📄 License

License: refers to the OpenPose's one.