Milestone Systems announced a forthcoming generative AI-powered video analytics plug-in for its XProtect® video management software, developed in collaboration with NVIDIA.
The new solution aims to help operators contextualize alarms and focus on what truly matters by automating video review, filtering out false alarms, and—based on early results—reducing operator alarm fatigue by up to 30%.
A beta version will debut at Smart City Expo World Congress in Barcelona (November 4–6), with general availability scheduled later this year.
Making Sense of More Video, Faster
Today’s video systems capture vast amounts of data, yet reviewing footage remains time-consuming and largely manual.
Milestone’s new XProtect plug-in leverages generative AI to automatically summarize, contextualize, and validate video content in real time—helping teams respond faster, manage video more efficiently, and minimize false alarms.
Key capabilities include:
- Automated Incident Reports: Selected video clips are converted into summaries and structured reports, reducing documentation time.
- Event Validation: Analyzes motion events and validates alarms to reduce false positives, fully integrated with the XProtect rule engine.
- Contextual Bookmark Summaries: Automatically generates natural-language summaries of bookmarked footage, allowing faster triage without manual review.
The plug-in integrates seamlessly with the XProtect rule engine and supports both on-premises and cloud deployments, ensuring flexibility and compliance across multiple industries.
Built on Ethical AI and Real-World Data
This new solution runs on Milestone’s Hafnia Vision Language Model (VLM) — trained on 75,000 hours of ethically sourced, real-world video data from Europe and the U.S. The data is prepared using NVIDIA Cosmos Curator, and the model operates on infrastructure powered by NVIDIA Cosmos Reason VLM.
“With this new XProtect plug-in, we are making advanced video intelligence accessible to cities, organizations, and operators everywhere who manage traffic systems,” said Thomas Jensen, CEO of Milestone Systems.
“It marks a pivotal step in our mission to transform how the world manages and learns from visual data — responsibly and at scale.”
Cities like Genoa, Italy, and Dubuque, Iowa, are among the first to explore the beta, demonstrating the real-world benefits of generative AI in public safety and traffic management.
Enabling Ecosystem Innovation with VLM-as-a-Service
The plug-in marks just the beginning. Milestone is also introducing VLM-as-a-Service (VLMaaS) via APIs, enabling developers, integrators, and partners to build custom generative AI solutions—regardless of their video management platform.
Live demonstrations will be held with Vaidio at Smart City Expo World Congress in Barcelona, hosted at the Dell booth, showcasing real-time incident summarization and AI model benchmarking tools.
Milestone will continue the conversation at the Milestone Developer Summit in Copenhagen (November 10–11), unveiling the winners of the Hafnia Hackathon and showcasing new developer innovations.
Milestone Systems is a global leader in data-driven video technology, serving industries such as manufacturing, airports, law enforcement, retail, and traffic management. Its solutions — including XProtect®, BriefCam®, and Arcules® — empower organizations to learn from the past, understand the present, and anticipate the future.
Milestone Systems Official Website
Frequently Asked Questions (FAQs)
1. What is the new Milestone Systems AI plug-in?
It’s a generative AI-powered video analytics plug-in for XProtect that automates video review, validates alarms, and summarizes incidents to save operator time.
2. How does it help reduce alarm fatigue?
By using AI to verify events and summarize video data, it can cut false alarms and streamline alert handling — reducing operator fatigue by up to 30%.
3. What is the Hafnia Vision Language Model (VLM)?
It’s Milestone’s proprietary AI model, trained on 75,000 hours of ethically sourced video data, designed for advanced video understanding and summarization.
4. Who are the key partners in this initiative?
The project is developed in collaboration with NVIDIA, Dell, and Vaidio, leveraging NVIDIA’s Cosmos AI framework for scalable performance.
5. When will the plug-in be available?
The beta is showcased at Smart City Expo World Congress 2025, with a full release expected later this year.