All Projects → webrtcHacks → facetouchmonitor

webrtcHacks / facetouchmonitor

Licence: other
monitor how often you touch your face with WebRTC and tensorflow.js

Programming Languages

javascript
184084 projects - #8 most used programming language
HTML
75241 projects
CSS
56736 projects

Projects that are alternatives of or similar to facetouchmonitor

pigallery
PiGallery: AI-powered Self-hosted Secure Multi-user Image Gallery and Detailed Image analysis using Machine Learning, EXIF Parsing and Geo Tagging
Stars: ✭ 35 (-45.31%)
Mutual labels:  tfjs
splat
Motion-controlled Fruit Ninja clone using Three.js & Tensorflow.js
Stars: ✭ 84 (+31.25%)
Mutual labels:  tfjs
color-pop
🌈 Automatic Color Pop effect on any image inspired by Google Photos
Stars: ✭ 21 (-67.19%)
Mutual labels:  bodypix
awesome-tensorflow-js
Awesome TensorFlow.js - A curated list of dedicated resources to master TensorFlow.js
Stars: ✭ 129 (+101.56%)
Mutual labels:  tfjs
Face-Mask
Real time webcam face detection, protect yourself from COVID19 with a virtual mask
Stars: ✭ 64 (+0%)
Mutual labels:  tfjs
dark-mode-clap-extension
Chrome extension to toggle dark mode on Netlify by clapping hands 👏
Stars: ✭ 82 (+28.13%)
Mutual labels:  tfjs
aio-rek
Face recognition based attendance system
Stars: ✭ 19 (-70.31%)
Mutual labels:  tfjs
nanodet
NanoDet: Tiny Object Detection for TFJS and NodeJS
Stars: ✭ 19 (-70.31%)
Mutual labels:  tfjs
stackml-js
Machine Learning platform in-browser for creators
Stars: ✭ 34 (-46.87%)
Mutual labels:  tfjs
parallax-effect
🤹🏻‍♂️ Parallax effect in javascript using face tracking. An immersive view in 3d with webcam.
Stars: ✭ 299 (+367.19%)
Mutual labels:  tfjs
backscrub
Virtual Video Device for Background Replacement with Deep Semantic Segmentation
Stars: ✭ 691 (+979.69%)
Mutual labels:  bodypix
Face Api.js
JavaScript API for face detection and face recognition in the browser and nodejs with tensorflow.js
Stars: ✭ 13,258 (+20615.63%)
Mutual labels:  tfjs
Tensorspace
Neural network 3D visualization framework, build interactive and intuitive model in browsers, support pre-trained deep learning models from TensorFlow, Keras, TensorFlow.js
Stars: ✭ 4,515 (+6954.69%)
Mutual labels:  tfjs

FaceTouchMonitor.com

A website that monitors when you touch your face.

In addition to washing your hands often, the CDC recommends that you avoid touching your eyes, nose, and mouth with unwashed hands. This can be hard to do without a reminder.

This project uses the BodyPix project from tensorflow.js to detect hands and faces. When a hand intersects with the face it is considered a touch. Set beep alerts or get a browser notification when you do. The reminders can you train yourself to stop accidental face touching.

No information is shared externally - no images are transmitted and all ML processing is done locally in the browser.

Try it at https://facetouchmonitor.com. Read about it at https://webrtchacks.com/

Demo video

It only takes seconds to try yourself.

Alt text

Usage Notes (and to do list)

Browsers and Devices

  • Browser support - tensorflow.js works best on Chrome and has been tested in the new Edge and Firefox. I haven't been able to make it work with Safari.
  • It will work on faster mobile devices, but the page is not responsive

Detection

  • The detection method is relatively crude - it has no depth perception, so a hand in front of your face will set it off.
  • A face is considered anything with one eye and a nose (according to its detection)
  • It usually does not detect as well if you are very close to the camera.
  • It is not optimized to handle multiple people

CPU Utilization

  • BodyPix can be CPU intensive on some devices - it will not work well on slower devices
  • If the FPS drops below 5 FPS, try using the Fastest setting
  • It works best if you can keep the browser tab in the foreground so background throttling is not invoked

Pull Request

The CSS fixes from dharmadeveloper108 are much appreciated!

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].