Coral: Expanding the edge AI landscape
summary
Coral presents concrete hardware and software updates: an Edge TPU that runs at four trillion operations per second, USB and M.2 accelerators (including a dual‑Edge TPU M.2), a Coral Dev Board Mini and an announced Dev Board Micro for tiny battery‑powered devices, TensorFlow Lite support, 25 precompiled models and Edge TPU runtime v13. Examples include care.ai patient monitoring, water‑meter analytics, and a restaurant grill system that improves cook timing; one 12‑module build reaches 48 TOPS for high‑throughput inference.
AI-written summary · about
chapters (7)
- 0:00Intro
- 2:15What Coral can do... Coral in action: high performance on-device ML inferencing at the edge
- 4:44Innovation with Al/ML at the edge Coral has inspired a huge wave of innovations of on-device Aland enabled many new products and services across al industry sectors from thousands of companies
- 7:04Coral M.2 modules
- 9:06New Coral Hardware Coral Dev Board Mini
- 10:50Coral Partnership program
- 11:32What's new coming up in 2021.. Coral Dev Board Micro
related talks
45mIntroducing Google Coral: Building On-Device AI
06mTop AI and ML announcements
47mProduction ready EdgeAI with Eclipse ioFog and Google Coral
16mIntroducing Coral Dev Board Micro
50mML and AI at the edge with Google Coral, EdgeTPU and Cloud Platform
55mEdgeAI - AI in the Palm of Your Hand
09mWelcome + Opening Remarks, Program Committee Members
59mBuilding models for TPU devices