A new processor from Renesas RZ/V has built-in vision AI accelerator

This new chip powered by two 64-bit Arm Cortex-A53 processor cores has a proprietary low power DRP-AI (Dynamically Reconfigurable Processor) accelerator engine to process vision AI at 1 TOPS/W (tera operations per second, per watt). RZ/V2MA features data-interfaces such as Ethernet, USB, and PCI Express to feed image-data input from multiple external cameras. By having both DRP-AI accelerator and OpenCV accelerator, this chip can perform accurate image recognition for machine vision systems such as AI powered gateways, video servers, security gates, POS terminals and robotic arms. To develop vision AI system faster Renesas is offering DRP-AI TVM based on the open-source deep learning compiler Apache TVM technology. DRP-AI TVM compiler allows DRP-AI accelerator work together with the CPU, so that DRP-AI can convert and generate more AI models. Renesas also supports ONNX and PyTorch AI models and in future to support Tensorflow too. Renesas is also offering "Vision AI Gateway Solution," a platform to detect and recognise objects in the pictures captured from multiple cameras. "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, 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 ex...