Explore the evolution and significance of AI at the edge in this 57-minute tinyML Talk from France. Delve into Anna Petrovicheva's, CTO of OpenCV.AI, insights on developing long-lasting, high-quality computer vision algorithms for battery-powered devices. Learn about the history and importance of edge computing, problem-solving approaches, data collection strategies, continuous learning techniques, and the comparison between HLS and HDL. Gain valuable knowledge on software and hardware considerations, tooling, and optimization methods like quantization and pruning for low-power vision applications.
Overview
Syllabus
Introduction
Why move to edge devices
History of edge devices
Questions
How to solve problem
Data bound problem
Pipeline
Data collection
Corrective feedback
Continuous learning
HLS vshdl
Software and hardware
Tooling
Quantization and pruning
Sponsors
Taught by
tinyML