ML at Client Side
summary
Concrete examples of on device ML running entirely in the browser using TFJS with WebAssembly, WebGL, WebGPU backends are demonstrated including image classification, pose estimation and body part segmentation that detect masks and people without sending data to servers. Models run in real time on typical laptops and phones (tens of frames per second), use small curated datasets and browser APIs for camera and sensors, and are deployable as pure JS or Node.js packages with measurable latency and accuracy tradeoffs.
AI-written summary · about
chapters (5)
- 0:00Introduction ML at client side
- 4:09Advantages of client-side ML
- 8:50Use cases and applications
- 13:20Model deployment and updates
- 17:26Summary and call to action
related talks
15mThe future of machine learning using JavaScript
46mFrom ML to LLM: on-device AI in the browser
29mSuperpowers for next gen web apps: Machine learning
41mTensorFlow.js 101: ML in the Browser and Beyond
24mThe Future of Machine Learning & Javascript
23mThe new era of ML in JavaScript
20mFrom ML to LLM: On-Device AI in the Browser
36mML at the Heart of JS