A 3D sensor tracks people and objects in large spaces. See how AI compute changes real-time analytics and smart infrastructure.
Tracking and analyzing people or objects across large spaces can be slow and inaccurate with conventional sensors, especially when AI models demand more compute. Organizations need a solution that delivers precise real-time analytics while handling demanding workloads.
Xovis’ PF-Series sensor meets this need. It combines a new SoC with an advanced 3D optical system to support large-area tracking and modern convolutional neural networks (CNNs). The sensor delivers up to 7.9 TOPS of CNN performance on an AI-native NPU, accelerating the convolutional models that power Xovis’ 3D tracking precision and optical pipeline. Its NPU runs up to three times faster than the widely used NXP i.MX8 and up to 22.5 times faster than the PC2SE-Series. A quad-core CPU paired with a 44.8 GFLOPS GPU provides high compute density for real-time processing and next-generation AI challenges.
Built on an AI-native architecture optimized for modern CNNs, the PF-Series delivers strong performance for CNN-based 3D vision while enabling future AI model innovations. By combining its optical system with a CNN-ready compute platform, it provides a scalable, high-accuracy foundation for spatial analytics in retail, mobility, smart buildings, and other smart infrastructure applications.
“Today marks a new era in retail analytics,” said Christian Studer, Xovis CEO and Co-Founder, speaking at NRF 2026 in Manhattan. “Xovis has put years of research and engineering into developing new sensor hardware with exceptional coverage and a powerful architecture. The research again proved that 3D stereovision outperforms other technology in responsibly measuring people flow in large environments.”
“We’ve witnessed tremendous advances in compute capabilities in recent years. Xovis has transformed those into a market-ready sensor that supports retailers’ transition to smarter, more automated processes,” said Florian Eggenschwiler, Xovis CPO.
