Apppublic

f1th4p/yolo11x-pose

sourceHugging Faceupdated 4mo agoView on Hugging Face

0likes

App README

YOLO11x-pose Space (Docker + FastAPI)

Free, always-on YOLO11x-pose endpoint for the health app's form pipeline.

POST /pose   {"image": "<base64 jpeg>"}
→ {"keypoints": [{"n":"nose","x":..,"y":..,"c":..}, ...], "w":..., "h":...}
GET  /        → {"status":"ok"}

Keypoints are COCO-17, normalized, y=0 at the bottom (matches form_coach.run_yolo_pose).

python

import base64, requests
b64 = base64.b64encode(open("frame.jpg","rb").read()).decode()
r = requests.post("https://<user>-yolo11x-pose.hf.space/pose", json={"image": b64})
print(r.json())

Free tier is CPU (~2-5s/frame) — this offloads the Mac, it isn't faster than its MPS GPU. See DEPLOY.md.