Life on the Edge
summary
Claims that 50 billion devices by 2020 drive moving processing to the edge and fog to cut bandwidth and improve privacy Security and cost motivate several concrete builds: OpenCV on Intel Joule running Nodejs performs real time face boxes and local classifiers, Mycroft and CMU Sphinx enable offline voice STT, Google AIY kit shows local keyword actions, gateways run MQTT and a gateway aggregation algorithm to batch data and add security, prototypes use Raspberry Pi Edison ESP8266 and a greenhouse demo that uses sensors wake every 6–8 hours and threshold wakeups to save battery.
AI-written summary · about
chapters (20)
- 0:00Intro
- 0:35Abstract
- 1:26Fog Cloud
- 2:09Cloud Computing
- 3:35Compute Power
- 4:26Costners Law
- 4:45Cloud Storage
- 5:08Privacy
- 6:52Computer Vision
- 7:49OpenCD
- 8:15Training Models
- 10:40Voice Recognition
- 11:54Open Source Models
- 12:39CMU Sinks
- 13:33Google Such Voice
- 14:31Gateways
- 16:27Gateway stack
- 17:38Project exploration
- 19:43Resources
- 28:04References
related talks
49mEnable IoT with Edge Computing and Machine Learning
39mExploring the Trend Toward the Edge
17mKeynote: Connecting the Dots on All Things Great and Small
52mLiving on the Cloud's Edge
07mLightning Talk: Living on the Edge - Computing as the Next Battleground
38mThe World Has Gone Edge – The Future of IoT and Smart Devices
41mCloud Computing: The Edge talks - long road into the Cloud
36mWebAssembly-based Edge AI Sensing Pipeline Orches...