Advanced Computing in the Age of AI | Saturday, December 9, 2023

AI Edge Partners Unveil Video Analytics Server 

via Shutterstock

Hailo, the AI chip startup, is teaming with a Japanese systems designer and manufacturing giant Foxconn to develop an AI edge processor aimed at video analytics applications.

The edge partnership is based on Foxconn Technology’s BOXiedge platform that integrates the Hailo-8 deep learning processor with a parallel processor from Japanese system-on-chip designer Socienext Inc. The combination creates a local video management server designed to shift workload processing from the cloud to the edge.

Yokohama-based Socionext specializes in low-power edge processors, including the recent release of an AI processor based on deep neural network technology. The approach is said to be “quantized,” that is, the device is tuned to specific values such as signals, rather than a continuous range of values.

The partners said Tuesday (May 12) the video analytics effort will be built around the AI chipmaker’s Hailo-8 deep learning processor designed to be embedded in edge devices. The neural network-based chip architecture is billed as enabling edge devices to handle deep learning applications that previously ran only in the cloud.

Processing data at the edge rather than in cloud datacenters is promoted as boosting efficiency and reducing latency for tasks like factory inspections and quality control. Another advantage is cost savings on data “egress” charges typically included in a cloud operator’s invoice.

The combined platform is geared toward standalone AI inference nodes and applications like industrial Internet of Things and healthcare along with smart cities and retail use cases. Those applications often handle multiple video streams that must be processed locally for minimal latency. Traffic monitoring and inspection cameras on factory floors are among the examples.

In one scenario, the edge platform would retrain models using various data sets to improve the accuracy of video analytics, including image classification and object segmentation.

Hence, Foxconn (TPE: 2354) said the AI edge platform is capable of processing and analyzing more than 20 streaming video feeds in real time.

“This collaboration will impact a wide range of industries increasingly driven by edge technology,” said Orr Danon, Hailo’s CEO and co-founder. “A new generation of chips means a new generation of capabilities at the edge.”

Hailo’s early customers include NEC (TYO: 6701), which previously announced it would use Hailo’s deep learning chip for “intelligent video analytics” applications.

Hailo’s AI chip is based on a proprietary design dubbed the “Structure-Defined Dataflow Architecture” that tops out at 26 tera operations per second. The chip architecture enablesedge devices to run deep learning applications that could previously only run on the cloud.

Hailo also claims its AI chip outperforms other edge processors in terms of size, performance and reduced power consumption.

As of March, Tel Aviv-based Hailo has raised $88 million in venture funding.

About the author: George Leopold

George Leopold has written about science and technology for more than 30 years, focusing on electronics and aerospace technology. He previously served as executive editor of Electronic Engineering Times. Leopold is the author of "Calculated Risk: The Supersonic Life and Times of Gus Grissom" (Purdue University Press, 2016).