Socionext Announces AI Accelerator for Edge Computing
Yokohama, Japan-based Socionext Inc. this week announced what it's calling the Neural Network Accelerator (NNA) engine, which is designed for deep-learning projects, has been "optimized for AI processing on edge computing devices."
"When implemented, it can achieve 100x performance boost compared with conventional processors for computer vision processing such as image recognition," the company stated.
Socionext said it plans to release the NNA later this year, along with an SDK (which will support TensorFlow) and related system-on-a-chip (SoC) products, which is what Socionext is known for. Socionext currently has SoCs on the market for Internet of Things (IoT), high-end graphics, high-speed networking, and visual and audio solutions, among others.
"Target applications [for the NNA] include automotive, in which high-precision object detection of images captured from car cameras, such as pedestrians and bicycles, can be used for driver assistance or automatic parking," the company commented in its announcement of the product.
"Another important application is display systems, such as TV and signage, in which the NNA boosts image recognition for super-resolution processing and helps high-definition imaging for 4K/8K screens."
Edge computing is a method of distributed computing designed to achieve increased real-time performance and bypass bottlenecks by having the processing take place as close to where the data is stored as possible, instead of first moving the data to, for example, a centralized cloud location. It is often seen as an ideal solution for high-data projects with intense and time-sensitive processing needs because of the decrease in latency the architecture promises.
More about the Socionext can be found here. More details about the product will be released in coming months.
Becky Nagel is the vice president of Web & Digital Strategy for 1105's Enterprise Computing and Education Groups, where she oversees the front-end Web team and deals with all aspects of digital strategy for the groups. She also serves as executive editor the ECG Web sites, and you'll even find her byline on PureAI.com, the ECG group's newest site for enterprise developers working with AI. She recently gave a talk at a leading technical publishers conference about how changes in Web technology may impact publishers' bottom lines. Follow her on twitter @beckynagel.