Zensors

Zensors provides physical AI for understanding real world spaces. Its platform processes video, sensor data and text from existing cameras and other devices, combines them with metadata from data warehouses into one aligned format, and uses multimodal AI to monitor how people, buildings and machines interact. AI agents then recommend or automate operational decisions so that operations teams can act instead of deciding what to do. The core model is Momenta, described as a large multimodal pre trained transformer built for video. It uses temporal attention across frames, fuses text and video features for open set object detection on new categories, applies semantic temporal tracking that follows meaningful attributes of objects rather than only their location, and understands higher level events and interactions. An active learning approach is meant to adapt it to new scenes. The vendor says it can replace several point solutions such as lidar, beacons and traditional computer vision products, works accurately on day one without extensive retraining, and is designed to preserve the anonymity of individuals. The site targets three sectors: aviation and transit, retail, and large commercial facilities. It highlights a deployment at Harry Reid International Airport in Las Vegas. Use cases listed include operational efficiency, strategic planning, financial performance, and safety and sustainability. Pricing is not published. Prospective customers are asked to request a demo or contact the company.