Imagination looks to add low-cost AI capabilities on edge devices

1 min read

Imagination Technologies has unveiled two neural network cores, the AX2185 and AX2145, that have been designed to enable high-performance computation of neural networks at very low power consumption in a minimal silicon area.

The cores are based on Imagination’s neural network accelerator (NNA) architecture, PowerVR Series2NX, which, according to the company, will help to bring ‘smartness’ from the cloud into edge devices, enabling greater efficiency and real-time responsiveness.

The Series2NX AX2185 targets the high-end smartphone, smart surveillance, and automotive markets, where neural network acceleration has a significant impact in areas such as image categorisation and driver assistance systems. Featuring eight full-width compute engines the AX2185 provides 2,048 MACs/clock (4.1 Tera Operations Per Second), which according to Imagination, represents the highest performance per mm2 in the market.

Optimised for cost-sensitive devices, the AX2145 targets mid-range smartphone, DTV/set-top box, smart camera and consumer security markets, which are increasingly adopting neural network acceleration for various tasks such as image management and vision-based applications.

The streamlined architecture delivers performance-efficient neural network inferencing for ultra-low bandwidth systems, making it a cost-effective solution for OEMs and ODMs working with limited silicon area budgets.

Both cores fully support the Android Neural Networks API (NNAPI), used by developers to bring neural network capabilities to Android-based mobile devices.

The PowerVR Series2NX architecture was designed from the ground-up to provide hardware acceleration for efficient neural network inference in mobile and embedded platforms. Its flexible bit-depth support on a per-layer basis for weights and data means PowerVR Series2NX can maintain high inference accuracy while reducing bandwidth/power requirements. It is, currently, the only solution supporting bit-depths from 16-bit to 4-bit, resulting in higher performance at lower bandwidth and power.

Imagination is also providing a comprehensive set of tools to streamline AI application development and deployment and, debugging and analysis. Machine learning frameworks are supported using our Network Developer Kit (NDK).

Russell James, vice president, Vision and AI, PowerVR, Imagination Technologies, said: “AI is transforming industries and neural networks are at the centre of this, creating exciting new use cases that are changing lives. The processing for this has predominantly taken place in the cloud, but with latency issues, privacy concerns and increasing demands on scalability, edge AI processing is becoming a necessity.”

The AX2185 core has already been delivered to lead customers and both cores are available for licensing now.