Introducing Google Coral - Building On-Device AI

Explore Google Coral, a new platform for on-device AI application development, in this 45-minute conference talk from Google I/O'19. Dive into the world of machine learning acceleration with TensorFlow demos and discover how Coral enables private, fast, and efficient neural network acceleration directly on your device. Learn about the technical specifications of Edge TPU hardware and software tools, as well as the application development process. Gain insights into key factors, components, and features of Coral, including the Dev Board, GPIO, deployment options, and Linux support. Understand the concept of transfer learning and its implementation through coding examples and a Teachable Machine demo. Explore real-world applications and workflows, and get inspired by user cases and potential contest opportunities. Whether you're a beginner or an experienced developer, this talk provides valuable information to help you grow your AI application ideas from prototype to production using Google Coral.

Introduction
Why Coral
Key Factors
What is Coral
Coral Components
Coral Devboard
HTTPU
Mobile Net Performance
Dev Board
GPIO
Deployment
Linux support
Hardware support
Software components
Python API
Image Classification
What is Transfer Learning
Coding Transfer Learning
Teachable Machine Demo
User Case
Workflow
Gravitylink
Applications
Contest