Adding Conditional Control to Text-to-Image Diffusion Models
Paper • 2302.05543 • Published • 58
from diffusers import ControlNetModel, StableDiffusionControlNetPipeline
controlnet = ControlNetModel.from_pretrained("phucleh2/controlnet_mask2cloth")
pipe = StableDiffusionControlNetPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", controlnet=controlnet
)Mask + Prompt → Cloth
- This model is still in development for better realistic-ness of the output. Stay tuned 🤗
- This model is a fine-tuned version of ControlNet, tailored to utilize a black-and-white outline of an upper garment (mask) and a descriptive prompt to generate a garment that aligns with the mask's outline.
- It's important to note that this model exclusively operates with upper garments. Please refrain from inputting masks for pants or jeans, as this could yield unexpected outcomes.
- Input mask size: 384 x 512
- Each time the model is executed, even with the same mask and prompt, it will generate a distinct output.
- This model is a fine-tuned of ControlNet using VITON-HD dataset
- Downloads last month
- 6
Model tree for phucleh2/controlnet_mask2cloth
Base model
runwayml/stable-diffusion-v1-5