tensorart/Bokeh_Pose_Controlnet

Bokeh_Pose_Controlnet

Description

• Input Image: Images processed by openpose/dwpose detection
• Output Image: Generated images with pose control

Openpose enables precise human body structure control, demonstrating better control effects and generalization compared to sd1.5 cn in our tests, and can easily adapt to lora models

Example

input	output	Prompt
		1 girl , thinking
		a man
		a young woman in room,wear a brown short,golden short hair
		a man in room,wear a brown vintage shirt,1990s

Use

We recommend using ComfyUI for local inference

With Bokeh

import torch
from diffusers import StableDiffusion3ControlNetPipeline
from diffusers import SD3ControlNetModel
from diffusers.utils import load_image

controlnet = SD3ControlNetModel.from_pretrained("tensorart/Bokeh_Openpose_Controlnet")
pipe = StableDiffusion3ControlNetPipeline.from_pretrained(
    "tensorart/bokeh_3.5_medium",
    controlnet=controlnet
)
pipe.to("cuda", torch.float16)

control_image = load_image("https://huggingface.co/tensorart/Bokeh_Pose_Control/resolve/main/images/001_pose.png")
prompt = "A woman thinking"
negative_prompt ="anime,render,cartoon,3d"
negative_prompt_3=""

image = pipe(
    prompt, 
    num_inference_steps=30,
    negative_prompt=negative_prompt, 
    control_image=control_image, 
    height=1728,
    width=1152,
    guidance_scale=4,
    controlnet_conditioning_scale=0.8
).images[0]
image.save('image.jpg')

Contact

• Website: https://tensor.art https://tusiart.com
• Developed by: TensorArt
• Api: https://tams.tensor.art/