ComfyUI-PoseKeypointsToImage

A ComfyUI custom node that renders pose images from POSE_KEYPOINT (DWPose/OpenPose format).

日本語

Description

This node generates pose images (IMAGE) from DWPose or OpenPose JSON format keypoint data. By passing pose_kps that has been filtered with FaceKeypointsFilter, you can create pose images with specific parts excluded.

Installation

Using ComfyUI Manager (Recommended)

Search for "PoseKeypointsToImage" in ComfyUI Manager and install.

Manual Installation

1. Navigate to your ComfyUI's custom_nodes folder
2. Clone this repository:

   git clone https://github.com/smartvain/ComfyUI-PoseKeypointsToImage.git
   

3. Restart ComfyUI

Usage

1. Connect the output of DWPose Preprocessor (or any node that outputs POSE_KEYPOINT) to this node
2. Configure the parameters:
   • width / height: Output image size (default: 512x512)
   • point_radius: Keypoint drawing radius (default: 2)
   • draw_body: Whether to draw body keypoints
   • draw_face: Whether to draw face keypoints
   • draw_hands: Whether to draw hand keypoints
3. Connect the output to other nodes

Node Location

Pose/Face Utils > Pose Keypoints → Image

Compatibility

• Works with DWPose Preprocessor from comfyui_controlnet_aux
• Compatible with OpenPose JSON format

License

MIT License