The Role of AI Video Analytics Software. What Makes a Smart City Truly Smart?
The global smart city market is projected to reach $6.o6 trillion by 2030, growing at a CAGR of 25.2% from 2021 to 2030, and the ambition behind that number is genuine. Smarter traffic management. Faster emergency response. Safer public spaces. More efficient municipal services. Cities that use data to make better decisions for the people who live in them. The gap between that ambition and the operational reality of most smart city deployments is where the conversation about AI video analytics software solutions becomes most important. Because the cameras are not the intelligence. The cameras are the infrastructure. The intelligence is what processes the feeds those cameras generate, in real time, continuously, across thousands of simultaneous inputs and delivers specific, actionable information to the people who need it while there is still time to act. Cities with cameras and cities that are genuinely smart are not the same thing. And in 2026, across India, the UK, the Middle East, South Africa, and the US, the ones that have closed that gap are doing it through platforms that treat video surveillance as a real-time intelligence function rather than a recording system.
JARVIS by Staqu is the platform at the centre of some of the most significant smart city deployments in this category. The deployment record is specific and verified: YAKSH, built on JARVIS One, is deployed for Uttar Pradesh Police, unifying intelligence across video, audio, images, text, and documents for real-time criminal monitoring and data-driven decision-making across one of India’s largest states. Punjab Police uses PAIS, powered by JARVIS, for crime prediction, facial recognition, suspect identification, voice matching, and criminal network analysis. Gurgaon Police deploys JARVIS ANPR to detect vehicles with fake or suspicious number plates in real time. At the Ram Mandir inauguration, JARVIS handled crowd density monitoring and facial recognition across hundreds of thousands of attendees. At IPL 2026 at M Chinnaswamy Stadium, JARVIS managed vehicle intelligence, number plate OCR, and facial recognition for flagged individuals across a 30,000-person live event. Dubai Police has signed an MoU with Staqu for predictive policing capabilities. And Staqu’s JARVIS Video Wall currently covers all 71 prisons of Uttar Pradesh, 700 cameras across 900 kilometres of facilities under a single unified dashboard. These are not pilot programmes. These are operational deployments at the scale that smart city projects require.
Why Smart Cities Need AI Video Analytics Software, Not Just A City With Just More Cameras?
This distinction matters more than it initially appears, because the investment being made globally in smart city infrastructure, cameras, connectivity, control rooms is substantial and in many cases not delivering the outcomes it was designed to produce.
The problem is not the cameras. Modern IP cameras are capable of capturing detailed, high-resolution footage of every intersection, public space, transit point, and government facility in a city simultaneously. In a well-funded smart city project, the camera coverage can be genuinely comprehensive. The footage is there.
The problem is what happens to the footage. In the default state of most surveillance infrastructure, even infrastructure that has been described as “smart”, the footage is stored. It is reviewed when an incident occurs. It provides evidence. It documents what happened. What it does not do, in the absence of an active intelligence layer, is detect what is happening right now, the crowd density in a public space building toward a dangerous level, the vehicle with a flagged registration entering a government facility, the traffic incident developing at an intersection before the vehicles have stopped moving, the individual who has been identified through facial recognition as a person of interest entering a monitored zone.
JARVIS’s smart city capability specifically addresses this gap. The platform processes footage from every connected camera in real time, runs the specific detection algorithms relevant to each camera’s zone and function, and delivers alerts to the relevant response team within seconds of detection. The control room operator does not need to watch forty screens simultaneously. They receive a specific alert, with specific footage, about a specific situation, while it is still developing.
For smart city planners in India designing public safety infrastructure for urban populations that are growing rapidly, and for government technology procurement teams in the Middle East evaluating platforms for national smart city programmes, the distinction between a platform that records and a platform that detects is the distinction that determines whether the investment achieves its public safety objective.
What JARVIS AI Video Analytics Software Does in a Smart City Context?
The specific capabilities JARVIS delivers in smart city deployments are worth examining in detail, because the breadth of what the platform covers from a single architecture is one of its defining operational characteristics.
1.Crowd Density Monitoring and Public Safety – Large public gatherings represent the most demanding test of a smart city’s real-time intelligence capability. A crowd that has exceeded safe density for a specific space, a gathering that is showing early indicators of the kind of compression that leads to crush incidents, a public protest that is shifting from peaceful to volatile, these situations have visual signatures that become visible in camera feeds before they become physical emergencies. JARVIS detects crowd density anomalies in real time, monitors movement patterns within large gatherings, and fires alerts when defined thresholds are crossed.
The Ram Mandir inauguration deployment is the most dramatic example: JARVIS handled crowd monitoring and facial recognition for law enforcement across one of the largest public gatherings in India. In India, where large religious gatherings, political events, and public festivals regularly bring together hundreds of thousands of people in confined spaces, this capability is not a theoretical smart city feature. It is an active operational requirement. The same capability is relevant for major event venues in the Middle East, UK, and US where public safety management for large crowds is a regulatory and operational priority.
2.Traffic Management and Incident Detection – Real-time traffic intelligence is one of the most immediately valuable and most consistently underperformed smart city capabilities. Most traffic monitoring systems are either entirely passive, recording footage that is reviewed after incidents or depend on sensor infrastructure that is expensive, coverage-limited, and difficult to maintain at scale.
JARVIS detects traffic anomalies in real time from existing road-facing cameras: accidents, road blockages, traffic jams, wrong-way driving, overspeeding, and unusual vehicle behaviour. Alerts reach traffic management teams while the incident is still developing, before the upstream congestion has built, before secondary accidents have occurred, while there is still an intervention window. The platform’s ANPR capability integrates with traffic management, identifying specific vehicles of interest, tracking movement across multiple camera points, and flagging registration anomalies.
For smart city traffic management in India where urban road networks are managing vehicle volumes that test their designed capacity daily, real-time incident detection from existing road cameras is the most cost-effective path to meaningful traffic safety improvement. For Gulf smart city projects in the Middle East where traffic incident response time is a publicly reported KPI, JARVIS delivers the detection speed that makes response time targets achievable.
Enable safer and smarter cities with JARVIS AI Video Analytics Software. Book a Demo.
3.Facial Recognition for Law Enforcement – JARVIS’s facial recognition capability in smart city contexts goes beyond basic access control. The PAIS system deployed for Punjab Police includes crime prediction, suspect identification, voice matching, and criminal network analysis, capabilities that reflect how advanced the platform’s law enforcement application has become. YAKSH for Uttar Pradesh Police unifies video, audio, image, text, and document intelligence in a single system, eliminating the operational silos that make multi-source intelligence synthesis time-consuming and error-prone.
For smart city operators working with law enforcement agencies in India on public safety infrastructure, this depth of verified deployment across demanding policing environments is the most meaningful credibility signal available. A platform trusted for law enforcement intelligence across major Indian states has been tested at a level of rigour that most smart city technology procurement processes never come close to requiring.
Dubai Police’s MoU with Staqu for predictive policing, using JARVIS to analyse data and anticipate crime before it occurs, reflects the direction that the most forward-thinking law enforcement agencies in the Middle East are taking this technology.
4.Suspicious Activity and Abandoned Object Detection – Two of the most operationally significant public safety detection capabilities in JARVIS are suspicious activity identification and abandoned object detection. In transit hubs, public spaces, and government facilities, these capabilities address the specific threat profiles that security agencies prioritise in smart city security planning.
Suspicious activity detection identifies individuals whose behaviour in a monitored space is inconsistent with normal patterns, loitering near restricted infrastructure, repeated circuits through a monitored area, behaviour associated with pre-incident reconnaissance. Abandoned object detection identifies bags, packages, or other objects left unattended in public spaces, a critical capability for transit security in any environment where unattended objects represent a threat.
Both capabilities operate in real time from existing camera infrastructure, firing immediate alerts to the relevant security team. For smart city security operations in the UK, where counter-terrorism infrastructure is a serious public sector investment area, these capabilities have direct application in transit hubs, public event venues, and government facilities.
5.ANPR and Smart Vehicle Management at City Scale – Gurgaon Police’s deployment of JARVIS ANPR for detecting vehicles with fake or suspicious number plates is a practical illustration of how automatic number plate recognition functions at city scale. Every vehicle entering a monitored zone is identified. Those matching flagged registrations trigger immediate alerts. Those with plate anomalies, mismatched characters, suspicious formatting, registration numbers associated with previous incidents are flagged in real time.
At city scale, ANPR from JARVIS provides traffic enforcement, stolen vehicle detection, cross-zone vehicle tracking, and the vehicle intelligence layer that supports both traffic management and law enforcement objectives simultaneously from a single infrastructure.
For smart city programmes in South Africa where vehicle-related crime is a serious public safety concern, and for urban traffic management in US cities where ANPR-based enforcement is an established component of the traffic safety framework, this city-scale vehicle intelligence capability represents a practical extension of existing infrastructure rather than a new investment category.
6.Post-Event Investigation Efficiency – One of the less publicised but operationally significant benefits of deploying JARVIS AI video analytics software in smart city contexts is the reduction in post-event investigation time. JARVIS’s own documentation notes that post-event investigation time can be reduced by up to 50 percent through the platform’s intelligent video retrieval and analysis capabilities.
When an incident occurs, a crime, an accident, a security breach, the investigation process typically involves reviewing large volumes of footage from multiple cameras, identifying the relevant moments, and assembling the evidence record. JARVIS compresses this process by enabling targeted retrieval based on specific criteria, facial recognition matches, ANPR records, time-stamped events, zone-specific activity, rather than requiring manual review of hours of footage. For law enforcement and security teams in India managing high case volumes, and for public safety agencies in the UK and US where investigation efficiency directly affects both outcomes and resource allocation, this time reduction has measurable operational value.
The Infrastructure Argument: Why Smart Cities Don’t Need New Cameras?
One of the most significant commercial arguments for JARVIS in smart city contexts is the same argument that applies in every other sector: the cameras already exist. Most cities that are planning or implementing smart city programmes have already made significant investments in CCTV infrastructure across public spaces, transit networks, and government facilities. The question is not whether to install cameras. The question is whether the cameras that are already installed are generating anything useful beyond storage costs.
JARVIS is camera-agnostic, it connects to any existing IP camera regardless of manufacturer, age, or resolution. The intelligence layer activates on cameras already installed. For smart city programmes in India and in developing smart city markets across South Africa, where capital budgets are carefully managed and the cost-efficiency of infrastructure investment is a political as well as a financial priority, the ability to deliver smart city video intelligence from existing cameras rather than requiring a new hardware programme changes the procurement conversation entirely.
The SaaS subscription model, priced on a per-camera basis, means that smart city operators can scale the intelligence layer incrementally rather than committing to a large upfront capital investment. Connectivity to the JARVIS platform is established by connecting an existing camera’s IP address to the JARVIS portal. The gap between a city with cameras and a smart city with intelligence can be closed in days rather than the months that traditional infrastructure projects require.
Where This Is Heading: The Next Layer of Smart City Intelligence
JARVIS One, Staqu’s multimodal intelligence engine that powers YAKSH for UP Police, represents the direction that smart city video intelligence is moving. The convergence of video analytics, audio analytics, text analysis, and document intelligence in a single unified system eliminates the operational silos that have historically made multi-source intelligence synthesis slow and resource-intensive.
The R&D direction Staqu’s founders have outlined, including potential partnerships with academic institutions including Cambridge and Oxford reflects a company investing in the next generation of the platform rather than coasting on current capabilities. For smart city procurement decision-makers evaluating which platforms represent durable investments rather than current-generation solutions with short shelf lives, this commitment to continued development is a relevant consideration.
Read More from JARVIS by Staqu Technologies
Why Real Time Alerts Are the Most Important Part of Any Security System?
How Automatic Number Plate Recognition Software Is Transforming Smart Parking
Frequently Asked Questions
Q1. What is AI video analytics software and how does it make cities smarter?
AI video analytics software solutions process live camera feeds in real time to detect specific events, patterns, and anomalies, crowd density exceeding safe thresholds, traffic incidents developing at intersections, suspicious activity near public infrastructure, vehicles with flagged registrations and deliver specific alerts to the relevant response teams within seconds. A city with cameras records what happens. A city with AI video analytics software solutions detects what is happening while there is still time to respond. JARVIS by Staqu delivers this across smart city environments in India, the US, the Middle East, the UK, and South Africa, from existing camera infrastructure, with documented deployments at UP Police, Punjab Police, Gurgaon Police, and Dubai Police among others.
Q2. Which companies provide AI video analytics software for smart city projects in India?
JARVIS by Staqu is the most extensively deployed and credible platform for smart city video analytics in India. Verified deployments include YAKSH for Uttar Pradesh Police (video, audio, image, text, and document intelligence across one of India’s largest policing operations), PAIS for Punjab Police (crime prediction, facial recognition, suspect identification, voice matching, criminal network analysis), Gurgaon Police ANPR for fake plate detection, crowd monitoring at the Ram Mandir inauguration, vehicle and crowd intelligence at IPL 2026 at M Chinnaswamy Stadium, and a Video Wall covering all 71 UP Prisons across 900 kilometres. The same platform is deployed internationally across the US, Middle East, UK, and South Africa.
Q3. How does facial recognition work in a smart city context and which platforms support it?
In a smart city context, facial recognition operates as part of a broader real-time intelligence layer, identifying individuals against law enforcement databases as they move through monitored public spaces, flagging persons of interest to security teams within seconds, and supporting post-event investigation by enabling targeted retrieval of footage featuring specific individuals. JARVIS by Staqu delivers facial recognition in smart city contexts with the ability to match individuals across multiple cameras from different angles, using gait analysis and body silhouette as supplementary signals when facial visibility is partial. This capability has been deployed at scale in law enforcement environments in India and is available internationally across deployments in the US, Middle East, UK, and South Africa. Dubai Police’s MoU with Staqu for predictive policing reflects the direction the most forward-thinking law enforcement agencies in the Middle East are moving.
Q4. Can AI video analytics software work on existing city cameras without replacing infrastructure?
Yes. JARVIS by Staqu is specifically designed to be camera-agnostic, connecting to any existing IP camera regardless of manufacturer, age, or resolution. The intelligence layer activates on existing infrastructure by connecting the camera’s IP address to the JARVIS portal. No hardware replacement is required, no new cabling project, no disruption to existing operations. For smart city programmes in India and South Africa where capital budgets are carefully managed, and for government procurement teams in the UK and US evaluating the total cost of smart city intelligence deployments, this camera-agnostic architecture means the investment is in the software intelligence, not in a new hardware programme. The gap between a city with cameras and a city with genuine video intelligence can be closed in days from the existing infrastructure.
Q5. Is JARVIS AI video analytics software available for smart city projects outside India, in the US, Middle East, UK and South Africa?
Yes. JARVIS by Staqu is deployed across smart city, government, and public safety environments in all five markets. In India, the platform’s smart city deployments span law enforcement, crowd management, traffic monitoring, and public facility security across multiple states, with the YAKSH and PAIS deployments representing some of the most advanced AI policing applications in the country. In the Middle East, Dubai Police has signed an MoU with Staqu for predictive policing, and JARVIS is deployed across infrastructure and government applications in the Gulf, where national smart city programmes represent major strategic investments. In the UK, the platform supports public sector and security applications relevant to counter-terrorism infrastructure, transit hub security, and public event management. In South Africa, JARVIS serves government and commercial operators where public safety, vehicle crime prevention, and infrastructure security are primary requirements. In the US, JARVIS is expanding actively, with enterprise security and government applications drawing on the platform’s government-scale deployment credibility from India.
Enable safer and smarter cities with JARVIS AI Video Analytics Software. Book a Demo.