CEVA EXPANDS AI ECOSYSTEM FOR ITS NPU IP WITH NEW PARTNERSHIPS FOR AUTOMOTIVE AND EDGE AI

Ceva announced that it has expanded the AI ecosystem for its class-leading NeuPro-M NPU IP with the addition of two new partnerships targeting automotive and vision-based Edge AI applications.

Visionary.ai
Visionary.ai empowers cameras to capture video in extreme low-light and HDR scenarios, with results that surpass human vision. Traditionally, low light and nighttime video has been one of the greatest imaging challenges, often resulting in dark, grainy, and unclear video. This new technology changes the game, enabling cameras to deliver unparalleled low-light performance. The technology can be deployed on the NeuPro-M NPU to enhance multiple Edge AI vision applications for a range of markets, including mobile, PC, IoT, robotics, automotive, security, smart cities, and medical imaging.

ENOT.ai
ENOT.ai specializes in neural network optimization, revolutionizing autonomous driving and Advanced Driver-Assistance Systems (ADAS). The collaboration combines Ceva's NeuPro-M NPU with ENOT.ai's revolutionary AI assistant technology, aimed at enhancing automotive safety, efficiency, and driver experience. Key features of this collaboration include an AI assistant that provides real-time lane departure warning, pedestrian detection, and voice instructions for vehicle operation, enhancing safety for new drivers and personalizing settings for an optimized driving experience. This partnership promises to set new standards in AI-driven automotive solutions, focusing on intuitive driver assistance, vehicle efficiency, and intelligent edge computing capabilities.

Ceva's NeuPro-M NPU IP addresses the processing needs of Classic and Generative AI with industry-leading performance and power efficiency for AI inferencing workloads. The NeuPro-M architecture is highly versatile and future-proof thanks to an integrated VPU (Vector Processing Unit), supporting any future network layer, and offers high scalability with four NPUs – the NPM11, NPM12, NPM14, and NPM18 to suit any AI workload.