Ceva will launch NeuPro artificial intelligence (AI) processors at CES 2018 (9 to 12 January). The family of processors is designed for deep learning inference at the edge.
The NeuPro family of processors is designed for smart and connected edge device vendors and builds on the company’s experience in deep neural networks for computer vision applications. Customers already deploy the Ceva-XM4 and Ceva-VA-XM6 vision platforms and CDNN, Ceva’s neural network software framework, in consumer, surveillance and ADAS products. This new family of dedicated AI processors offers a considerable step-up in performance, explains Ceva, ranging from two Tera operations per second (TOPS) for the entry-level processor to 12.5TOPS for the most advanced configuration.
The NeuPro processor line extends the use of AI beyond machine vision to new edge-based applications including natural language processing, real-time translation, authentication, workflow management, and many other learning-based applications that make devices smarter and reduce human involvement.
Ilan Yona, vice president and general manager of the Vision Business Unit at CEVA, commented: “It’s abundantly clear that AI applications are trending toward processing at the edge, rather than relying on services from the cloud. The computational power required along with the low power constraints for edge processing, calls for specialised processors rather than using CPUs, GPUs or DSPs. We designed the NeuPro processors to reduce the high barriers-to-entry into the AI space in terms of both architecture and software. Our customers now have an optimised and cost-effective standard AI platform that can be utilised for a multitude of AI-based workloads and applications.”
The NeuPro architecture combines hardware- and software-based engines. Optimisations for power, performance, and area (PPA) are achieved using a precise mix of hardware, software and configurable performance options for each application tier, says Ceva.
There are four AI processors in the NeuPro family, offering different levels of parallel processing. The NP500 is the smallest processor, and targets IoT, wearables and cameras. It has 512 MAC units. The NP1000 includes 1024 MAC units and targets mid-range smartphones, ADAS, industrial applications and AR/VR headsets. The NP2000 includes 2048 MAC units and targets high-end smartphones, surveillance, robots and drones. Finally, the NP4000 includes 4096 MAC units for high-performance edge processing in enterprise surveillance and autonomous driving.
Each processor consists of the NeuPro engine and the NeuPro VPU. The NeuPro engine includes the hardwired implementation of neural network layers which include convolutional, fully-connected, pooling, and activation.
The NeuPro VPU is a cost-efficient programmable vector DSP, which handles the CDNN software and provides software-based support for advances in AI workloads. NeuPro supports both eight- and 16-bit neural networks. The MAC units achieve better than 90 per cent utilisation when running, ensuring highly optimised neural network performance. The overall processor design reduces DDR bandwidth substantially, improving power consumption levels for any AI application, adds Ceva.
The NeuPro family, coupled with CDNN, provides deep learning for developers to easily and efficiently generate and port their proprietary neural networks to the processor. CDNN supports the full gamut of layer types and network topologies, enabling fastest time-to-market, assures Ceva.
In conjunction with the NeuPro processor line, Ceva will also offer the NeuPro hardware engine as a convolutional neural network (CNN) accelerator. When combined with the CEVA-XM4 or CEVA-XM6 vision platforms, this provides a flexible option for customers looking for a single, unified platform for imaging, computer vision and neural network workloads.
NeuPro will be available for licensing to select customers in Q2 2018 and for general licensing in Q3 2018.
