Sign In

Communications of the ACM

ACM TechNews

Stanford Engineers Create New AI Camera for Faster, More Efficient Image Classification

View as: Print Mobile App Share:
Researchers say the hybrid optical-electrical computer could be ideal for autonomous vehicles.

Researchers at Stanford University have devised a new type of artificially intelligent camera system that can classify images faster and more energy efficiently.

Credit: Andrey Suslov/Getty Images

Stanford University researchers have developed an artificial intelligence camera system that classifies images faster and more efficiently.

They achieved this by combining two types of computers into one, creating a hybrid optical-electrical computer designed specifically for image analysis.

The first component of the prototype camera is an optical computer that does not require the power-intensive mathematics of digital computing, while the second component is a traditional digital electronic computer.

The optical computer physically preprocesses image data, filtering it in multiple ways that an electronic computer would otherwise have to do mathematically. This layer operates with zero input power, saving the hybrid system time and energy and resulting in significantly fewer calculations, fewer calls to memory, and less time to complete the process.

The prototype rivals existing electronic-only computing processors in both speed and accuracy, but with substantial computing cost savings.

From Stanford News
View Full Article


Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA


No entries found