Milestone Systems has announced the development of a generative AI-powered video analytics plug-in for its XProtect® video management software, created in partnership with NVIDIA.
The new tool, designed to aid operators in prioritising alarms and concentrating on significant tasks, promises to automate video review processes and potentially lessen operator alarm fatigue by up to 30%, according to preliminary results.
The initial beta version is set to debut at the Smart City Expo World Congress in Barcelona from 4-6 November, with a full release expected later in the year.
Enhancing video review efficiency
Traditionally, reviewing captured video footage is a laborious task, despite the vast quantity of data collected. Milestone Systems' XProtect plug-in addresses this challenge by employing generative AI to summarise, contextualise, and verify video content in real-time.
This facilitates quicker responses and more efficient video management, significantly aiming to reduce operator alarm fatigue.
The plug-in offers several key features, including:
- Automated Incident Reports – Instantly transforms selected video clips into concise incident summaries and structured reports, allowing operators to spend less time on documentation.
- Event Validation – Configurable to assess motion events and authenticate alarms, minimising false alarms and enhancing alert management. This capability is seamlessly integrated into the XProtect rule engine.
- Contextual Bookmark Summaries – Uses natural language to automatically summarise bookmarked footage, enabling quick triage without the need for manual review of each clip.
Flexible deployment options
Integrating directly with the XProtect rule engine, the plug-in can be installed either on-premises or in the cloud, providing versatility in compliance and deployment options.
It is built on Milestone’s Hafnia Vision Language Model (VLM), which has been trained on 75,000 hours of real-world video data sourced ethically from Europe or the US, with NVIDIA Cosmos Curator utilised for data preparation. The platform runs on cloud infrastructure or regional data centres powered by NVIDIA technology.
CEO commentary
XProtect users and partners have a unique opportunity to enhance their capabilities
Thomas Jensen, CEO of Milestone Systems, stated: “With this new XProtect plug-in, we are making advanced video intelligence accessible to cities, organisations, and operators everywhere who manage traffic systems – helping them unlock new levels of efficiency, safety, and insight.”
He also emphasised the transformative potential of this advancement, noting that XProtect users and partners have a unique opportunity to enhance their capabilities. Cities like Genoa, Italy, and Dubuque, Iowa, are among the XProtect customers eager to explore these new features to improve traffic management systems.
Expanding ecosystem capabilities
Beyond the plug-in, Milestone is introducing a Vision Language Model as a Service via APIs. This allows developers, integrators, and partners to create their own generative AI solutions, independent of the video management platform being utilised.
Exhibitions of the XProtect plug-in will occur at the Smart City Expo World Congress, showcasing a new AI model benchmarking tool and live incident summarisation in partnership with Vaidio.
Looking ahead
The momentum will continue at the Milestone Developer Summit in Copenhagen on 10-11 November, where the full capabilities of Hafnia VLM will be demonstrated, alongside the announcement of the Hafnia Hackathon winners.
Milestone Systems, a world pioneer in data-driven video technology, announced a forthcoming generative AI-powered video analytics plug-in for its XProtect® video management software, developed in collaboration with NVIDIA.
Designed to help operators contextualise alarms and focus on what truly matters, the new tool automates video review, filters out false alarms, and based on initial findings could reduce up to 30% of operator alarm fatigue.
A beta version will debut at Smart City Expo World Congress in Barcelona, November 4-6, with general availability coming later this year.
Making sense of more video, faster
The video systems capture vast amounts of data, yet reviewing footage remains time-consuming and largely manual. Milestone Systems’ new XProtect plug-in addresses this challenge by leveraging generative AI to automatically summarise, contextualise, and validate video content in real time, helping teams respond faster, manage video more efficiently, and effectively reduce operator alarm fatigue.
Key capabilities include:
- Automated Incident Reports – Selected video clips are instantly converted into incident summaries and structured reports, helping operators reduce time spent on documentation.
- Event Validation – The plug-in can be configured to analyse motion events and validate alarms, reducing false positives and improving alert handling. This feature is fully integrated with the XProtect rule engine.
- Contextual Bookmark Summaries – Bookmarked footage is automatically summarised using natural-language output, allowing fast triage without reviewing each clip manually.
The plug-in integrates directly with the XProtect rule engine and is deployable on-premises or in the cloud to support compliance and deployment flexibility.
Built on ethical AI, powered by real-world data
This new solution is built on Milestone’s Hafnia Vision Language Model (VLM) trained on 75,000 hours of ethically sourced, real-world video data from either Europe or the US, using NVIDIA Cosmos Curator for data preparation and running either on cloud infrastructure or regional data centres powered by NVIDIA.
It leverages the NVIDIA Cosmos Reason VLM, making it one of the most advanced and compliant video AI platforms in the industry.
Advanced video intelligence
Thomas Jensen, CEO of Milestone Systems, said: “With this new XProtect plug-in, we are making advanced video intelligence accessible to cities, organisations, and operators everywhere who manage traffic systems – helping them unlock new levels of efficiency, safety, and insight."
"XProtect users will get access to state-of-the-art generative AI capabilities, and our partners will be able to build value on top of those new capabilities now available within XProtect. It truly marks a pivotal step in our mission to transform how the world manages and learns from visual data, responsibly and at scale.”
XProtect customers like the cities of Genoa, Italy, and Dubuque, Iowa, are excited to try these new capabilities, leading the way in adopting advanced video intelligence solutions to enhance traffic management.
Enabling ecosystem innovation with VLM-as-a-Service
The plug-in is just the beginning. Milestone is also introducing a VLM as a Service via APIs, allowing developers, integrators, and partners to build their own generative AI solutions regardless of the video management platform in use.
Live demonstrations of the XProtect plug-in will be held in partnership with Vaidio at Smart City Expo World Congress on November 4-6 in Barcelona, in the Dell booth, showcasing a new AI model benchmarking tool and real-time incident summarisation.
Continuing the momentum at the Milestone Developer Summit
Milestone will continue the momentum at the Developer Summit in Copenhagen, Nov. 10–11, where Hafnia’s capabilities and the winners of the Hafnia Hackathon will be revealed.
