NVIDIA's latest AI Blueprint, announced ahead of the Smart City Expo World Congress, provides developers with tools to build visual AI agents that can analyse video streams and image content. The technology, part of NVIDIA Metropolis, combines computer vision and generative AI technologies to create customisable workflows.
The Blueprint is designed to help enterprises and public sector organisations develop AI agents that enhance workforces relying on visual information from cameras, IoT sensors, and vehicles. According to the announcement, these agents can answer user questions, generate summaries, and enable alerts for specific scenarios.
A key feature of the system is its user-friendly approach, allowing customisation through natural language prompts instead of rigid software code. The technology is powered by vision language models (VLMs), which combine computer vision and language understanding to interpret the physical world and perform reasoning tasks.
Early adopters are already implementing the technology across various sectors. In Southeast Asia, systems integrators ITMAX in Malaysia and FPT in Vietnam are building AI agents for smart city and intelligent transportation applications. In Italy, K2K is working with city traffic managers in Palermo to deploy visual AI agents using NIM microservices and NVIDIA AI Blueprints.
The technology offers practical applications across multiple sectors. In warehouses, AI agents can alert workers to safety protocol breaches. For infrastructure maintenance, workers can use AI agents to review aerial footage and identify degrading roads, train tracks, or bridges.
The Blueprint is available free for developers to experience and download, with production deployment possible through NVIDIA AI Enterprise, the company's end-to-end software platform for AI development.