custom-diffusion 0.1.2

Custom Diffusion: Creating Video from Frame Using Multiple Diffusion

Custom Diffusion: Creating Video from Frame Using Multiple Diffusion

Installation

pip install custom_diffusion

Usage

from custom_diffusion.pipelines.controlnet_pipeline import StableDiffusionControlNetGenerator


generator = StableDiffusionControlNetGenerator()

generated_image = generator.generate_image(
     stable_model_path="runwayml/stable-diffusion-v1-5"
     controlnet_model_path="lllyasviel/control_v11p_sd15_canny",
     scheduler_name="DDIM",
     image_path="test.png",
     prompt="Anime boy",
     negative_prompt="bad",
     height=512,
     width=512,
     guess_mode=False,
     num_images_per_prompt=1,
     num_inference_steps=20,
     guidance_scale=7.0,
     controlnet_conditioning_scale=1.0,
     generator_seed=0,
     preprocess_type="Canny",
     resize_type="center_crop_and_resize",
     crop_size=512,
)