Skip to main content

Custom Diffusion: Creating Video from Frame Using Multiple Diffusion

Project description

Custom Diffusion: Creating Video from Frame Using Diffusion

downloads pypi version HuggingFace Spaces

Installation

pip install custom_diffusion

Usage

# Importing the required libraries

from custom_diffusion.utils.data_utils import load_images_from_folder
from custom_diffusion import StableDiffusionControlNetGenerator
from custom_diffusion.utils.video_utils import convert_images_to_video
from custom_diffusion.demo import video_pipeline

# Creating a video from a video file
frames_path = video_pipeline(
    video_path="test.mp4",
    output_path="output.mp4",
    start_time=0,
    end_time=5,
    frame_rate=1,
)

# Creating a video from a folder of images
images_list = load_images_from_folder(frames_path)

prompt = "a anime boy"
negative_prompt = "bad"

list_prompt = [prompt] * len(images_list)
list_negative_prompt = [negative_prompt] * len(images_list)

# Generating images from a list of images
generator = StableDiffusionControlNetGenerator()

generated_image_list = generator.generate_image(
    stable_model_path="andite/anything-v4.0",
    controlnet_model_path="lllyasviel/control_v11p_sd15_canny",
    scheduler_name="EulerAncestralDiscrete",
    images_list=images_list,
    prompt=list_prompt,
    negative_prompt=list_negative_prompt,
    height=512,
    width=512,
    guess_mode=False,
    num_images_per_prompt=1,
    num_inference_steps=30,
    guidance_scale=7.0,
    controlnet_conditioning_scale=1.0,
    generator_seed=0,
    preprocess_type="Canny",
    resize_type="center_crop_and_resize",
    crop_size=512,
)

# Converting the generated images to a video
frame2video = convert_images_to_video(
    image_list=generated_image_list,
    output_file="output.mp4",
    frame_rate=5,
)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

custom_diffusion-0.1.9.tar.gz (17.0 kB view details)

Uploaded Source

File details

Details for the file custom_diffusion-0.1.9.tar.gz.

File metadata

  • Download URL: custom_diffusion-0.1.9.tar.gz
  • Upload date:
  • Size: 17.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.9.12

File hashes

Hashes for custom_diffusion-0.1.9.tar.gz
Algorithm Hash digest
SHA256 35609dd81f7dc8e5e1a0fe4e037cb0c3d9e1e9405afea57b02d14631f96f31c7
MD5 bb8068fa50e466b0c7a12587ec046530
BLAKE2b-256 2fc1946d80681467d18d824d0ae27b43c0ecbeecbf36a3ee4ce45de54fa2d4cb

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page