Machine learning on a chip developer SiMa.ai is partnering CVEDIA and Supermicro to develop smart city solutions for embedded edge systems. The companies will combine CVEDIA’s AI video analytics with Supermicro’s edge appliances, incorporating SiMa.ai’s Machine Learning System on Chip (MLSoC) platform.

The aim is to enable local municipalities to improve performance and cost of their applications such as object detection, anomaly detection, vehicle classification, and line crossing at the edge.

“SiMa.ai’s mission is to help customers deploy high-performance AI/ML applications at the edge with a high bar for energy-efficiency and ease of use with partners,” said Elizabeth Samara-Rubio, Chief Business Officer of SiMa.ai. “We are pleased to introduce real-time AI/ML partner solution integrations such as CVEDIA’s library of reliable, affordable AI-based video analytics solutions to the market that will help customers scale their AI/ML edge development challenges effortlessly with one platform.”

“Leveraging CVEDIA’s strength in Intelligent Video analytics and SiMa.ai‘s software centric MLSoC, we are excited to introduce an innovative edge AI/ML solution that strikes an optimum balance between performance and power,” said Arjan Wijnveen, CEO of CVEDIA. “The collaboration enables a completely new era of AI/ML solutions at the edge that optimise and scale the use cases, and propel value added services.”

SiMa.ai delivers a platform for edge AI that scales, from computer vision, to transformers to multimodal generative AI. The SiMa.ai MLSoC enables multiple AI models to run simultaneously on a single camera stream, with improved latency and accuracy. It is designed to provide increased compute capabilities to run multiple channels concurrently, while maximising efficiency by delivering the highest performance frames per second per watt (FPS/W).