Making Your Android Things Devices Smarter
summary
Hands-on details about making Android Things devices smart: a working wildlife camera uses a TensorFlow Lite image classifier to spot bear moose etc and sends alerts, ML Kit runs on-device or falls back to cloud with ML Kit labels: on-device ~400 vs cloud ~10,000 and supports an 80% confidence threshold, the Google Assistant embeds into hardware and assistant returns JSON responses for custom actions, and cloud Speech/Translate combo makes a translator like Pixel Buds.
AI-written summary · about
chapters (15)
- 0:00Start
- 0:53Image Recognition
- 1:56Tensor Flow
- 4:14Mlk for Android
- 5:17Predefined Models
- 5:39Labels Api
- 6:10Text Recognition Api
- 7:39Face Detection
- 8:06Landmark Detection Api
- 10:49Firebase Vision Detectors
- 11:39The Google Assistant
- 13:25Smart Home Integration
- 14:45Google Cloud Machine Learning
- 15:08Cloud Vision Api
- 16:13Sample Android Things Project
related talks
19mMaking Your Android Things Devices Smarter
21mLive Coding: Adding Machine Learning to Android Things in 10 Mins
43mWhat’s new in Android machine learning
58mWorking With Android Things: Machine Learning With TensorFlow
13mWhat's new in Android machine learning
47mMyths and Reality of Android Machine Learning
39mMachine Learning on Android Demystified
43mMyths and Reality of Android Machine Learning