Archives: Presentations

Quantization for Efficient Inference in Edge Devices

NeuroFabric: A Priori Sparsity for Training on the Edge

Memory-Optimal Direct Convolutions for Maximizing Classification Accuracy in Embedded Applications

Network Architecture Search for Efficient Wake-word Detection

ReBNet: Residual Binarized Networks

Resource-Constrained Keyword Spotting

TinyAI: from edgeAI tools to Neuro-Spiking Architectures

Developer Choices and Trends: Insights from the Embedded Vision Alliance’s Computer Vision Developer Survey

The Eye in IoT: a unique solution by Himax & emza that includes special CMOS sensor, custom made processor and lean ML algorithm

Deep Learning on Off-the-shelf Cameras with Examples in Top View, People Detection, Facial Recognition, and Car Detection