ComfyUI-OpenPose

ComfyUI-OpenPose
★ 37

人体姿态估计OpenPose关键点提取模型下载
在ComfyUI中集成OpenPose，自动下载模型、从单张图片提取人体关键点并输出可视化图与JSON。
💡 从单张图片提取并可视化人体关键点，导出JSON。
🍴 5 Forks💻 Python🔄 2024-07-31
📦
网盘下载
复制链接后前往夸克网盘下载
https://pan.quark.cn/s/8b0992d318c3
📦 requirements.txt
torch
opencv-python
numpy
huggingface_hub
torchvision
Example Workflow BODY_25
Example Workflow COCO
📄 README

OpenPose Node

This repository contains a Python implementation for extracting and visualizing human pose keypoints using OpenPose models.

The OpenPoseNode class allows users to input images and obtain the keypoints and limbs drawn on the images with adjustable transparency.

Additionally, it can provide an image with only the keypoints drawn on a black background, and the keypoints value in a JSON format.

Example

You can find the example workflow here

Features

  • Download OpenPose models from Hugging Face Hub and saves them on ComfyUI/models/openpose
  • Process imput image (only one allowed, no batch processing) to extract human pose keypoints.
  • Draw keypoints and limbs on the original image with adjustable transparency.
  • Generate an image with only the keypoints drawn on a black background.
  • Generate JSON with list of keypoints
  • Choose between two OpenPose representations: COCO and BODY_25.

    • Prerequisites

    Ensure you have the following packages installed:

  • torch
  • cv2 (OpenCV)
  • numpy
  • huggingface_hub
  • torchvision

    • you can install them with

  • pip install -r requirement.txt

    • or

  • python install.py

    • Acknowledgements

    Special thanks to beingjoey/pytorch_openpose_body_25 for the inspiration. Their work provided a foundation and motivation for developing this custom node.

    Contributing

    We welcome contributions! If you find any issues or have suggestions for improvements, please open an issue or submit a pull request.

    License

    This project is licensed under the GPL-3.0 License. See the LICENSE file for details.

    Thank you for using the OpenPose node for ComfyUI. We hope it enhances your image processing workflows!