This package has been deprecated

Author message:

Package no longer supported. Contact Support at https://www.npmjs.com/support for more info.

faceplugin

How it works

Installation

npm install faceplugin

Table of Contents

• Face Detection
• Face Landmark Extraction
• Face Liveness Detection
• Face Expression Detection
• Face Pose Estimation
• Eye Closeness Detection
• Gender Detection
• Age Detection
• Face Feature Embedding

Examples

https://user-images.githubusercontent.com/82228271/187199680-208c7995-21b1-48d7-8758-3977452ff955.mp4

• Vue.js Demo
• React.js Demo

[Please subscribe on Youtube channel]

Documentation

Here are some useful documentation

Face Detection

Load detection model

loadDetectionModel()

Detect face in the image

detectFace(session, canvas_id)

Face Landmark Extraction

Load landmark extraction model

loadLandmarkModel()

Extract face landmark in the image using detection result

predictLandmark(session, canvas_id, bbox)

Face Liveness Detection

Load liveness detection model

loadLivenessModel()

Detect face liveness in the image using detection result. (Anti-spoofing)

predictLiveness(session, canvas_id, bbox)

Face Expression Detection

Load expression detection model

loadExpressionModel()

Detect face expression

predictExpression(session, canvas_id, bbox)

Face Pose Estimation

Load pose estimation model

loadPoseModel()

Predict facial pose

predictPose(session, canvas_id, bbox, question)

Eye Closeness Detection

Load eye closeness model

loadEyeModel()

Predict eye closeness

predictEye(session, canvas_id, landmark)

Gender Detection

Load gender detection model

loadGenderModel()

Predict gender using face image

predictGender(session, canvas_id, landmark)

Age Detection

Load age detection model

loadAgeModel()

Predict age using face image

predictAge(session, canvas_id, landmark)

Face Recognition

Load feature extraction model

loadFeatureModel()

Extract face feature vector in 512 dimension

extractFeature(session, canvas_id, landmarks)

Contact

For any inquiries or questions, please contact us at [e-mail].