📦 Models

Each of the following models can be combined and reconfigured in real time.

Model: MediaPipe Hands (2D)

📚 MediaPipe Hands documentation

• 21 2D hand landmarks per hand
• Track up to 4 hands at once
• Pinching states, hand pointers, and gestures

Model: TensorFlow Handpose (3D)

📚 TensorFlow Handpose documentation

• 21 3D hand landmarks
• Can only track 1 hand at a time
• 📅 Extra helpers and plugins coming soon

Model: MediaPipe FaceMesh

📚 MediaPipe FaceMesh documentation

• 468 2D face landmarks
• Track up to 4 faces at once
• 📅 Extra helpers and plugins coming soon

Model: MediaPipe Pose

📚 MediaPipe Pose documentation

• Full body mode with 33 2D pose landmarks
• Upper body mode with 25 2D upper pose landmarks
• 📅 Extra helpers and plugins coming soon

Model: TensorFlow Handpose

📚 Jeeliz Weboji documentation

• 6DOF head pose estimations
• 11 face morphs and 16 helper states
• Comes with "Face Pointer" based plugins