Renesas boosts vision AI accelerator for multi-camera support

New Products |
By Nick Flaherty

Renesas Electronics has expanded the performance of its AI accelerator in its RZ/V microprocessor range using new compiler technology to support multiple cameras.

The RZ/V2MA uses the proprietary low power DRP-AI (Dynamically Reconfigurable Processor) accelerator alongside two 1GHz 64bit ARM Cortex-A53 cores with performance of 1 TOPS/W.

The RZ/V2MA makes use of recently acquired development tools to aid vision AI system design. In addition to the existing DRP-AI Translator, the new device adds DRP-AI TVM, based on the EdgeCortix MERA Compiler Framework and open-source deep learning compiler Apache TVM technology. Renesas teamed with EdgeCortix in July to port the technology.

While DRP AI Translator is designed to convert AI models into DRP-AI executables, the DRP-AI TVM compiler lets the DRP-AI accelerator work together with the CPU, allowing DRP-AI to convert and generate more AI models As a first phase, Renesas supports ONNX and PyTorch AI models and plans to support Tensorflow in the future.

The RZ/V2MA includes Ethernet, USB, and PCI Express interfaces for image input from multiple external cameras as well as video codec blocks for the H.265 and H.264 standards.

In addition to the DRP-AI accelerator, the RZ/V2MA includes an OpenCV accelerator that allows rule-based image processing simultaneously. These features bring highly accurate image recognition capabilities for machine vision products such as AI-equipped gateways, video servers, security gates, POS terminals and robotic arms.

“One of the challenges for embedded systems developers who want to implement machine learning is to keep up with the latest AI models that are constantly evolving,” said Shigeki Kato, Vice President of Renesas’ Enterprise Infrastructure Business Division. “With the new DRP-AI TVM tool, we are offering designers the option to expand AI frameworks and AI models that can be converted to executable formats, allowing them to bring the latest image recognition capabilities to embedded devices using new AI models.”

“Renesas’ RZ/V series is ideal for embedded devices since it does not need fans or heat sinks, due to its extremely low power consumption and low heat capability when running AI,” said Chiharu Nakabayashi, President of amnimo, a  provider of IoT and AI-based services that is a subsidiary of Yokogawa Electric. “With these devices, we are confident that we can develop powerful image AI gateways that can be installed anywhere.”

Renesas has developed the “Vision AI Gateway Solution,” which is an AI-based object detection and recognition platform that uses multiple cameras to collect and efficiently transmit data wirelessly. The reference design combines the RZ/V2MA MPU with other Renesas products such as power ICs, VersaClock clock generator, and communication modules for Wi-Fi, Bluetooth, and LTE.

The RZ/V2MA and development tools are available now in a 15 x 15mm GBA.

Other articles on eeNews Europe


Linked Articles
eeNews Europe