IKM® State of Art

With great enthusiasm, we are integrating a powerful framework that enables the creation, testing, and optimization of AI agents within SimReady digital twins, in conjunction with the NVIDIA Omniverse Blueprint for Smart Cities.

Real-time exploration of urban scenarios anticipates the impact of every decision, shaping dynamic city ecosystems. Infrastructure, mobility, and services are precisely modeled around the needs of the community.
Agentic AI and Physical AI transform the urban environment into a distributed, responsive mind capable of harmonizing metropolitan complexity with intelligence and sustainability.

NVIDIA and K2K® embrace this ambitious challenge, helping define a new paradigm: cities that think, learn, and evolve.
Smarter public spaces, more seamless services, and a daily life enriched by deep, sensitive interaction between people, environment, and innovation.


Thanks to this synergy, the urban fabric comes alive with intelligence and vision.

A breakthrough in visual AI at city scale has just been realized.

K2K® is one of the early adopters of the newly announced NVIDIA Omniverse Blueprint for Smart City AI. to optimize operations AI agents and digital twins for cities. The blueprint provides the complete software stack needed to efficiently develop AI agents and digital twins at scale.

K2K® has also successfully launched the curation of a vast urban video dataset to fine-tune next-generation Vision Language Models (VLMs) optimized for the NVIDIA AI Blueprint for video search and summarization (VSS) workloads. This milestone combines cutting-edge data curation infrastructure with powerful AI training services on NVIDIA DGX Cloud.

The curated data covers city life across a full set of diverse conditions: traffic density, lighting variations, weather events, and pedestrian flow, reflecting the rich complexity of real-world urban dynamics. The initiative lays the groundwork for intelligent systems capable of perceiving, reasoning, and interacting with the physical environment in real time.


From Raw Video to Real Intelligence


The core challenge in building high-performance vision AI agents lies not only in computing power, but in the quality and scale of AI training data available. This project addresses that by curating city-scale video footage using advanced pipelines designed for regulatory compliance, privacy-preserving transformations, and multimodal annotation.

Once curated, the video data is processed in DGX Cloud using fine-tuning microservices optimized for NVIDIA AI Blueprints. This stack accelerates the development of highly interactive, perceptive visual AI agents capable of powering a new class of smart city services, from real-time traffic summarization to multimodal urban analytics powered by K2K®’s Neural Networks Architectures.


Training Vision AI Agents for the Real World


This work enables a new generation of VLMs fine-tuned specifically for dense, live video environments. These models are built for low-latency video understanding and retrieval, key to real-time analysis such as live camera search, scene summarization, and generative events reporting.

Aligned with the NVIDIA AI Blueprint for Video Search and Summarization (VSS), the project demonstrates how purpose-trained VLMs can serve as foundational components for digital agents operating across complex dynamic physical spaces. Whether deployed at intersections, transportation hubs, or public venues, these agents are capable of answering questions, generating summaries, and surfacing critical information directly from video feeds.


DGX Cloud: A Platform for Scalable Urban AI


The initiative showcases the synergy of high-performance AI infrastructure and modular data engineering. DGX Cloud streamlined the model, fine-tuning with scalability and speed, eliminating the traditional bottlenecks of compute-bound training cycles. NVIDIA provides the backbone for compliant, high-throughput video ingestion, enrichment, and labeling—accelerating the path from raw footage to high-value AI-ready data.


A Foundation for the Next Phase of Urban Intelligence


This effort is the first of several strategic steps aimed at operationalizing sovereign AI for cities, architectures of Neural Networks that are not only powerful and accurate, but also auditable, explainable, and aligned with public values.

The NVIDIA Omniverse Blueprint for Smart City AI provides a framework that focuses on training AI to perceive and respond in complex scenarios, testing the AI in a digital twin, and deploying the AI through AI agents. Next, K2K® plans to explore the use of digital twins for simulating and testing AI agents before deployment.

The fine-tuned VLM emerging from this pipeline is designed to support a broad range of use cases, from traffic flow optimization to public safety insights, infrastructure resilience, and emergency response.

By abstracting visual reasoning into adaptable services, this initiative empowers both public and private stakeholders to unlock the full potential of their video assets, with full compliance, responsibly, and at scale.

Be Genetic

Learn how video Al agents analyze thousands of video streams in real time, leveraging NVIDIA GPUs and NVIDIA Al Blueprint for video search and summarization to transform complex data into actionable insights.

This session will showcase how Palermo is exploiting cutting-edge technology to drive the digital industrial revolution in urban settings, extracting, managing, and capitalizing on data to make cities smarter.


NVIDIA GTC2025, San Jose

Video GenAI is revolutionizing the way systems learn, adapt, and operate in real-time.

This technology opens up new possibilities in video stream analysis, extracting insights that were previously unreachable.

Thanks to advanced tools like Neural Ordinary Differential Equations (ODE) and State Space Models (SSM), it is possible to overcome complex challenges, unlocking the full potential of VLMs in real-world scenarios.

These innovations are transforming Smart Cities, making real-time video analysis a key resource for the future.


NVIDIA Webinar January 22, 2025

Smart City Expo World Congress 2024

Generative Neural Network Architectures
for the Management of Complex Processes,
the Genesis of Information.

K2K® is proud to announce that IKM® Neural Networks Architectures now include NVIDIA NIM microservices, part of the NVIDIA AI Enterprise software platform, and vision language models (VLMs) to power smart city applications with advanced video generative AI. This integration is set to elevate the neural networks' capabilities in traffic and waste management, public safety, and environmental analysis, enabling cities to efficiently manage resources and improve urban life.

1. IKM® Global VQA
IKM® Neural Networks, powered by NVIDIA NIM microservices, supercharge real-time analytics across multi-camera streams with high precision. City operations teams can obtain instant answers to natural language queries, as well as engage in multi-turn conversation spanning multiple data sources and correlate images to further refine query results, allowing for immediate and highly specific insights into issues such as roadworks, crowd density, and environmental hazards. This breakthrough capability enables faster, more adaptive decision-making in critical urban situations, which can transform the way cities respond to challenges.

2. OVCR® Traffic Management
The OVCR® Traffic Management system provides real-time vehicle tracking, congestion analysis, and speed analysis. This data allows city planners to optimize traffic flow, enhance road safety, and respond quickly to incidents. The neural networks’ capabilities also give traffic managers a complete, real-time overview of traffic patterns and road conditions, making cities safer and more efficient.

3. IKM® Waste Management
With NVIDIA NIM microservices, IKM® Neural Networks analyze video streams of waste bin fill levels and surrounding cleanliness. Real-time insights into illegal dumping or full waste bins allow cities to better plan waste collection and maintain cleaner environments, reducing operational costs and contributing to urban aesthetics.

K2K® is helping set a new benchmark for smart cities, harnessing the power of neural networks, VLMs, and NVIDIA AI Blueprints to create cleaner, safer urban environments.
K2K® is excited to drive the future of smart cities and deliver innovations that improve urban life for all residents using NVIDIA AI.

K2K® is empowering cities with advanced video search, summarization and video generative AI capabilities, powered by NVIDIA’s AI software stack and NVIDIA AIBlueprints, which are designed to tackle complex urban challenges by delivering real-time insights that help manage resources, streamline operations, and ensure public safety in urban environments.

NVIDIA AI Blueprints are reference workflows for generative AI use cases, built with NVIDIA NIM microservices as part of the NVIDIA AI Enterprise software platform. The new AI Blueprint for video search and summarization helps developers like K2K leverage NVIDIA Metropolis to create a diversity of AI agents that can perform visual understanding of activities within video and to build smart city services.

By integrating NVIDIA AI Blueprints, part of NVIDIA AI Enterprise, IKM® Technology brings real-time, intelligent insights to smart cities. Through state-of-the-art neural networks and video genAI, the IKM® Neural Networks Architectures can support various urban applications, from optimizing traffic flow to enhancing waste management and analyzing environmental conditions. These enhancements enable smart cities to make informed, knowledge-driven decisions for managing urban resources efficiently and effectively.

Users can specify prompts and events that the system will analyze and summarize, giving city operators immediate access to information on incidents like traffic blockages, large gatherings, and environmental concerns.

This not only allows for a quicker response to urban developments but also significantly boosts city management and situational awareness.

This collaboration with NVIDIA brings exceptional scalability and adaptability to IKM® Neural Networks, enabling cities to harness the power of real-time insights and alerts for a smarter and safer urban experience. K2K® is helping set new standards in smart city technology, making cities not only more connected but also more resilient and sustainable.

See K2K® in action at Smart City Expo World Congress in Barcelona, November 5-7th. Find us demoing in Dell Technology’s pavilion Gran Via, Hall 3, Level 0, Street D, Stand 101 and the AMSYS-Lenovo booth Gran Via, Hall 3, Level 0, Street C, Stand 41.

Generative physical Al NIM microservices - and NVIDIA Metropolis reference workflows - are helping create intelligent, immersive work environments.


July 29, 2024

Neural Networks, Video Knowledge Management & Management of Information


June 21, 2023

With the graphical processing units (GPUs), numerous algorithms have been adapted for parallel computing, achieving execution times that are approaching real time. Specifically for emotion recognition from facial images, these computational enhancements have ushered in the second class of approaches, neural networks — and in particular, Convolutional Neural Networks (CNNs).

AI includes all theories and techniques needed to develop algorithms that make computers capable of performing “intelligent” activities related to specific domains. So far, we are talking about a kind of decoding that can be performed by human beings as well.