Skip to content

Latest commit

 

History

History
44 lines (34 loc) · 1.94 KB

README.md

File metadata and controls

44 lines (34 loc) · 1.94 KB

jweb-face-landmarker

A self contained example demonstrating how to use MediaPipe FaceLandmarker with Max's jweb connected to either a live webcamera stream or using still images.

Max patcher example of jweb-face-landmarker

Features

MediaPipe FaceLandmarker provides 52 features (blendshapes) per face, which are mapped to live.slider objects in the example patcher.

Eyes browDownLeft browDownRight
Eyes browOuterUpLeft browOuterUpRight
Eyes cheekSquintLeft cheekSquintRight
Eyes eyeBlinkLeft eyeBlinkRight
Eyes eyeLookDownLeft eyeLookDownRight
Eyes eyeLookInLeft eyeLookInRight
Eyes eyeLookOutLeft eyeLookOutRight
Eyes eyeLookUpLeft eyeLookUpRight
Eyes eyeSquintLeft eyeSquintRight
Eyes eyeWideLeft eyeWideRight
Brow, Cheek browInnerUp cheekPuff
Nose noseSneerLeft noseSneerRight
Mouth jawLeft jawRight jawForward jawOpen
Mouth mouthClose mouthFunnel mouthPucker
Mouth mouthRollLower mouthRollUpper mouthShrugLower mouthShrugUpper
Mouth mouthDimpleLeft mouthDimpleRight
Mouth mouthFrownLeft mouthFrownRight
Mouth mouthLeft mouthRight
Mouth mouthLowerDownLeft mouthLowerDownRight
Mouth mouthPressLeft mouthPressRight
Mouth mouthSmileLeft mouthSmileRight
Mouth mouthStretchLeft mouthStretchRight
Mouth mouthUpperUpLeft mouthUpperUpRight

Notes

Still images seem to work best when faces are not too far from the camera.

Resources

This example is inspired by an example by Rob Ramirez, which is in turn inspired by MediaPipe in JavaScript.