Everything a Swift Dev Needs to Know About Machine Learning

on YouTube ↗22m

" summary

Company captures a short face video, runs computer vision + ML to fit a detailed 3D face model and then manufactures bespoke frames, so the pipeline actually produces real glasses. Model training happens in Python (TensorFlow) using pretrained nets like Inception v3 fine‑tuned on thousands of images; deployment options are server inference on AWS or on‑device inference with Metal Performance Shaders or by bundling TensorFlow. Rendering and AR compositing occur locally with SceneKit, currently rendering on phone after server inference.

AI-written summary · about

" chapters (7)

1. 0:00Introduction
2. 1:40What is Machine Learning
3. 4:30Neural Networks
4. 8:43Does this affect Swift
5. 11:44Can I use TensorFlow and Swift
6. 13:41Whats it like to work on ML
7. 18:12How to get started

" related talks

• 25mGetting Started with Machine Learning in your iOS appsMeghan Kane · App Builders Switzerland 2017
• 40mDeep Learning on iOSShuichi Tsutsumi · 360iDev 2017
• 33mSee More, Be MorePaolo Bianchi, Cristian Cascetta, Stefano Bodini · #Pragma Conference 2023
• 34mThe Potential of IMAGE Processing in Swift for DevelopersElina Semenko · Swift Heroes 2023
• 09mMachine Learning on the EdgeSangeetha KP · CodeLand 2020
• 11mAI & Deep Learning Using Core ML and Metal FrameworkAkansha Sharma · try! Swift India 2017
• 14mDo you want an easy way to add Machine Learning into your app?Sandra Dupré · Android Makers 2019
• 27mMachine Learning and MobileBrett Koonce · Scale By The Bay 2019