Making Your Android Things Devices Smarter
summary
Shows how to train and run TensorFlow Lite models on Android Things hardware, explains using ML Kit for on‑device and cloud image APIs (about 400 on‑device vs 10,000 cloud labels), and demonstrates concrete projects: a camera that classifies bears/moose to alert wildlife services, a 3D‑printed lamp controlled via Assistant traits, and a pork torii translation demo that actually translates speech. Includes implementation details (TF convert, 80% confidence threshold, Camera2 integration) and links to code on Hackster.
AI-written summary · about
chapters (6)
- 0:00Start
- 3:07Environment Setup
- 3:47Create TensorFlow Files
- 4:19Optimize for TensorFlow Lite
- 4:43Integrate with App
- 13:13Setup Traits
related talks
18mMaking Your Android Things Devices Smarter
58mWorking With Android Things: Machine Learning With TensorFlow
21mLive Coding: Adding Machine Learning to Android Things in 10 Mins
39mMachine Learning on Android Demystified
42mCreating Smarter Applications
29mFrom TensorFlow to ML Kit: power your Android application with ML
47mMachine Learning on Android Demystified
13mWhat's new in Android machine learning