SDPose: Whole-Body Pose Estimation
ComfyUI Wiki
SDPose provides whole-body pose estimation for ComfyUI, using RT-DETRv4 for person detection and SDPose for keypoint estimation.
SDPose is a whole-body pose estimation model for ComfyUI. It detects human body keypoints including face, hands, and body joints. The pipeline uses RT-DETRv4 for initial person detection, then SDPose for precise keypoint estimation.
Usage in ComfyUI
Place the model files as follows:
ComfyUI/models/checkpoints/—sdpose_wholebody_fp16.safetensorsComfyUI/models/diffusion_models/— RT-DETRv4 detection models
Requires a ComfyUI custom node that supports SDPose for loading and inference.
| Developer | Comfy-Org Community |
| Release Date | 2025 |
| Architecture | SDPose + RT-DETRv4 Detection |
| License | Apache-2.0 |
| Type | Pose Estimation / Keypoint Detection |
| Capabilities | Whole-body pose estimation (body + face + hands) |
Links
- Hugging Face (Comfy-Org): Comfy-Org/SDPose
Guides and workflows related to this model series.
No articles found.
Comments
Sign in with GitHub to join the discussion.