yinguobing/head-pose-estimation

Health v2 · maintenance only

Adoption not part of the score

1353 stars · 295 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A Python library for realtime human head pose estimation using ONNX Runtime and OpenCV. It combines face detection (SCRFD), 68-point facial landmark detection, and PnP-based pose calculation from pre-trained deep learning models.

Use cases

• estimate head pose from a webcam in realtime
• get head orientation angles from a video file
• detect facial landmarks with a deep learning model
• build driver drowsiness or attention monitoring
• run face pose estimation without GPU with ONNX Runtime
• estimate 3D head rotation from 2D face images

When to choose

• you need realtime head pose (pitch/yaw/roll) from webcam or video
• you want a lightweight CPU-friendly pipeline via ONNX Runtime
• you need facial landmarks plus pose in one ready-made solution

When to avoid

• you need full 3D face reconstruction or dense mesh tracking
• you need a production web service with an API out of the box
• you require PyTorch-native training or inference (training lives in a separate repo)

Facets

library · maturity stable

computer-vision image-processing machine-learning deep-learning computer-vision artificial-intelligence deep-learning python cross-platform head-pose-estimation facial-landmarks onnxruntime opencv pnp realtime face-detection linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="yinguobing/head-pose-estimation")