Real Time Face Detection through a Webcam Using AI
summary
Builds a working real‑time face detector that runs entirely in the browser using TensorFlow.js and face-api.js with a webcam stream, WebGL acceleration, and locally hosted model files. The demo loads a tiny face detector / MobileNet variant, detects faces with confidence scores, overlays 68 landmarks and classifies a small set of expressions (happy, sad, surprised, neutral). It runs in real time, takes about twenty minutes to assemble from examples, and successfully tracks and labels faces on screen.
AI-written summary · about
chapters (13)
- 0:00Introduction and speakers
- 1:16Talk agenda and survey
- 3:10Speaker background
- 4:20Brief history of facial recognition
- 11:29AI evolution and the modern era
- 17:12Shift to edge computing
- 22:32Key tools and libraries
- 25:15Live demo setup
- 32:10Real-time face detection demo
- 35:32Facial landmarks detection
- 39:58Emotion detection with AI
- 42:51Building and learning from the project
- 47:22Future of AI and Q&A
related talks
51mRealTime Exoskeleton-Mapping and Body Segmentation on the Browser with Tensorflow.js
56mTensorFlow.js: Make a smart webcam in JS with a pre-trained ML model
41mAI in Web: TensorFlow js
41mTensorFlow.js 101: ML in the Browser and Beyond
32mSelf-Portraits in JavaScript
32mWe Also Can Do It! Machine Learning in Javascript!
34mA Powerful Introduction to Machine Learning with Js
15mMachine learning for next gen web apps with TensorFlow.js