Abstract

Today's widespread video acquisition and interpretation applications [eg, autonomous driving (Beltrán et al., 2020), surveillance (Xie et al., 2021), object detection (Jiao et al., 2022), object tracking (Wu et al., 2021), and anomaly detection (Mansour et al., 2021)] are fueled by CMOS image sensors (CIS) and deep learning algorithms. However, these computer vision systems suffer from energy inefficiency and throughput bottlenecks (Chai, 2020) that stem from the transmission of a high volume of data between the sensors at the edge and processors in the cloud. For example, smart glasses (eg, Meta AR/VR glasses,

Metadata

publication

Physical neuromorphic computing and its industrial applications, 148, 2023

year

2023

publication date

2023/8/2

authors

Bert Offrein, Michael Wynn Hopkins, Leslie Samuel Smith, Md Abdullah-Al Kaiser, MAA Kaiser, G Datta, Z Wang, AP Jacob, PA Beerel, AR Jaiswal

link

https://books.google.com/books?hl=en&lr=&id=ocLOEAAAQBAJ&oi=fnd&pg=PA148&dq=info:r-Lu8tyysAQJ:scholar.google.com&ots=UM6ZlckW1Y&sig=ipuZH-y8qHwpnCDAsw0BaGAvBSU

journal

Physical neuromorphic computing and its industrial applications

pages

148

publisher

Frontiers Media SA