TinyML Talks - Low-Power Computer Vision

Explore low-power computer vision technologies in this 31-minute tinyML Talks webcast featuring Yung-Hsiang Lu from Purdue University. Discover energy-efficient solutions for battery-powered systems, including parameter quantization, compressed convolutional filters, network architecture search, and knowledge distillation. Learn about hierarchical neural networks for reducing energy consumption on embedded systems and get introduced to the IEEE International Low-Power Computer Vision Challenge. Gain insights into recent research, challenges, and open questions in the field of low-power computer vision.

Introduction
YungHsiang Lu
LowPower Computer Vision
Outline
Quantization
ImageNet
Network Architecture Search
Distillation
Questions
Data Sets
Benchmarks
Recent research
Challenges
Paper
Open Questions
CVD Challenge
LPC Vida
Summary
Question
Survey Paper
tinyML Talk Sponsors
Edge Impulse
Sincents
Next talk
Thanks