ComfyUI-MagicAnimate

★ 224

动画生成ComfyUI扩展自动安装Magic Animate

在ComfyUI中集成Magic Animate，便捷生成动画帧与序列，简化流程与依赖管理（需约15GB磁盘空间）。

💡 在ComfyUI中快速生成和管理动画帧序列。

🍴 24 Forks💻 Python🔄 2024-05-22

🔗 GitHub 原文

📦

网盘下载

复制链接后前往夸克网盘下载

https://pan.quark.cn/s/51137d50651f

📦 requirements.txt

diffusers
omegaconf
torchvision
Pillow

📄 README

ComfyUI-MagicAnimate

Easily use Magic Animate within ComfyUI!

[](https://discord.gg/MfVCahkc2y)

Installation

Node types

Example workflows

Animate any person’s image with a DeepPose video input

Need help? Join our Discord!

Installation (this custom node will auto-download ~15GB of model weights!)

Option 1: Automatic installation (Recommended)

Please install “ComfyUI-MagicAnimate” via the ComfyUI-Manager.

Option 2: Manual installation

cd ComfyUI/custom_nodes/
git clone https://github.com/thecooltechguy/ComfyUI-MagicAnimate
cd ComfyUI-MagicAnimate/
python -m pip install -r requirements.txt

Node types

MagicAnimateModelLoader

Loads the MagicAnimate model

MagicAnimate

Uses the MagicAnimate model to animate an input image using an input DeepPose video, and outputs the generated video

Example workflows

Animate any person’s image with a DeepPose video input

https://comfyworkflows.com/workflows/2e137168-91f7-414d-b3c6-23feaf704fdb

TODO

[x] Fix white color in output (credits: @mingqizhang)

[x] Optimize amount of models downloaded (now requires 15GB and only downloads the models needed)

[ ] Refactor to use any Model, VAE, and ControlNet input from ComfyUI (this should also further reduce number of models downloaded)