Deploying spatio-temporal forecasting models across many cities is difficult: traffic networks differ in size and topology, data availability can vary by orders of magnitude, and new cities may provide only a short history of logs. Existing deep traffic models are typically trained per city and backbone, creating high maintenance cost and poor transfer to data-scarce cities. We ask whether a single, backbone-agnostic layer can condition on "which city this sequence comes from", improve accuracy in full- and low-data regimes, and support better cross-city adaptation with minimal code changes. We propose CityCond, a light-weight city-conditioned memory layer that augments existing spatio-temporal backbones. CityCond combines a city-ID encoder with an optional shared memory bank (CityMem). Given a city index and backbone hidden states, it produces city-conditioned features fused through gated residual connections. We attach CityCond to five representative backbones (GRU, TCN, Transformer, GNN, STGCN) and evaluate three regimes: full-data, low-data, and cross-city few-shot transfer on METR-LA and PEMS-BAY. We also run auxiliary experiments on SIND, a drone-based multi-agent trajectory
The automated generation of interactive 3D cities is a critical challenge with broad applications in autonomous driving, virtual reality, and embodied intelligence. While recent advances in generative models and procedural techniques have improved the realism of city generation, existing methods often struggle with high-fidelity asset creation, controllability, and manipulation. In this work, we introduce CityGenAgent, a natural language-driven framework for hierarchical procedural generation of high-quality 3D cities. Our approach decomposes city generation into two interpretable components, Block Program and Building Program. To ensure structural correctness and semantic alignment, we adopt a two-stage learning strategy: (1) Supervised Fine-Tuning (SFT). We train BlockGen and BuildingGen to generate valid programs that adhere to schema constraints, including non-self-intersecting polygons and complete fields; (2) Reinforcement Learning (RL). We design Spatial Alignment Reward to enhance spatial reasoning ability and Visual Consistency Reward to bridge the gap between textual descriptions and the visual modality. Benefiting from the programs and the models' generalization, CityGen
Neural radiance fields (NeRF) and its subsequent variants have led to remarkable progress in neural rendering. While most of recent neural rendering works focus on objects and small-scale scenes, developing neural rendering methods for city-scale scenes is of great potential in many real-world applications. However, this line of research is impeded by the absence of a comprehensive and high-quality dataset, yet collecting such a dataset over real city-scale scenes is costly, sensitive, and technically difficult. To this end, we build a large-scale, comprehensive, and high-quality synthetic dataset for city-scale neural rendering researches. Leveraging the Unreal Engine 5 City Sample project, we develop a pipeline to easily collect aerial and street city views, accompanied by ground-truth camera poses and a range of additional data modalities. Flexible controls over environmental factors like light, weather, human and car crowd are also available in our pipeline, supporting the need of various tasks covering city-scale neural rendering and beyond. The resulting pilot dataset, MatrixCity, contains 67k aerial images and 452k street images from two city maps of total size $28km^2$. On
Two fundamental issues surrounding research on the image of the city respectively focus on the city's external and internal representations. The external representation in the context of this paper refers to the city itself, external to human minds, while the internal representation concerns how the city is represented in human minds internally. This paper deals with the first issue, i.e., what trait the city has that make it imageable? We develop an argument that the image of the city arises from the underlying scaling of city artifacts or locations. This scaling refers to the fact that, in an imageable city (a city that can easily be imaged in human minds), small city artifacts are far more common than large ones; or alternatively low dense locations are far more common than high dense locations. The sizes of city artifacts in a rank-size plot exhibit a heavy tailed distribution consisting of the head, which is composed of a minority of unique artifacts (vital and very important), and the tail, which is composed of redundant other artifacts (trivial and less important). Eventually, those extremely unique and vital artifacts in the top head, i.e., what Lynch called city elements,
The emergence of smart cities and sustainable development has become a globally accepted form of urbanization. The epitome of smart city development has become possible due to the latest innovative integration of information and communication technology. Citizens of smart cities can enjoy the benefits of a smart living environment, ubiquitous connectivity, seamless access to services, intelligent decision making through smart governance, and optimized resource management. The widespread acceptance of smart cities has raised data security issues, authentication, unauthorized access, device-level vulnerability, and sustainability. This paper focuses on the wholistic overview and conceptual development of smart city. Initially, the work discusses the smart city idea and fundamentals explored in various pieces of literature. Further various smart city applications, including notable implementations, are put forth to understand the quality of living standards. Finally, the paper depicts a solid understanding of different security and privacy issues, including some crucial future research directions.
3D city generation with NeRF-based methods shows promising generation results but is computationally inefficient. Recently 3D Gaussian Splatting (3D-GS) has emerged as a highly efficient alternative for object-level 3D generation. However, adapting 3D-GS from finite-scale 3D objects and humans to infinite-scale 3D cities is non-trivial. Unbounded 3D city generation entails significant storage overhead (out-of-memory issues), arising from the need to expand points to billions, often demanding hundreds of Gigabytes of VRAM for a city scene spanning 10km^2. In this paper, we propose GaussianCity, a generative Gaussian Splatting framework dedicated to efficiently synthesizing unbounded 3D cities with a single feed-forward pass. Our key insights are two-fold: 1) Compact 3D Scene Representation: We introduce BEV-Point as a highly compact intermediate representation, ensuring that the growth in VRAM usage for unbounded scenes remains constant, thus enabling unbounded city generation. 2) Spatial-aware Gaussian Attribute Decoder: We present spatial-aware BEV-Point decoder to produce 3D Gaussian attributes, which leverages Point Serializer to integrate the structural and contextual character
Does the national innovation city and smart city pilot policy, as an important institutional design to promote the transformation of old and new dynamics, have an important impact on the digital economy? What are the intrinsic mechanisms? Based on the theoretical analysis of whether smart city and national innovation city policies promote urban digital economy, this paper constructs a multi-temporal double difference model based on a quasi-natural experiment with urban dual pilot policies and systematically investigates the impact of dual pilot policies on the development of digital economy. It is found that both smart cities and national innovation cities can promote the development of digital economy, while there is a synergistic effect between the policies. The mechanism test shows that the smart city construction and national innovation city construction mainly affect the digital economy through talent agglomeration effect, technology agglomeration effect and financial agglomeration effect.
Smart cities have been a very active research area in the past 20 years, while continuously adapting to new technological advancements and keeping up with the times regarding sustainability and climate change. In this context, there have been numerous proposals to expand the scope of smart cities, focusing on resilience and sustainability, among other aspects, resulting in terms like smart sustainable cities. At the same time, there is an ongoing discussion regarding the degree in which smart cities put people at their centre. In this work, we argue toward expanding the current smart city definition by integrating the circular economy as one of its central pillars and adopting the term smart (and) circular city. We discuss the ways a smart and circular city encompasses both sustainability and smartness in an integral manner, while also being well-positioned to foster novel business activity and models and helping to place citizens at the heart of the smart city. In this sense, we also argue that previous research in smart cities and technologies, such as those related to Industry 4.0, can serve as a cornerstone to implement circular economy activities within cities, at a scale that
We propose a method to procedurally generate a familiar yet complex human artifact: the city. We are not trying to reproduce existing cities, but to generate artificial cities that are convincing and plausible by capturing developmental behavior. In addition, our results are meant to build upon themselves, such that they ought to look compelling at any point along the transition from village to metropolis. Our approach largely focuses upon land usage and building distribution for creating realistic city environments, whereas previous attempts at city modeling have mainly focused on populating road networks. Finally, we want our model to be self automated to the point that the only necessary input is a terrain description, but other high-level and low-level parameters can be specified to support artistic contributions. With the aid of agent based simulation we are generating a system of agents and behaviors that interact with one another through their effects upon a simulated environment. Our philosophy is that as each agent follows a simple behavioral rule set, a more complex behavior will tend to emerge out of the interactions between the agents and their differing rule sets. By c
This chapter explores the six core dimensions of smart cities (i.e. smart economy, mobility, environment, people, living, and governance) emphasizing their interdependence and the need for holistic orchestration. Building on Giffinger et al. (2007) and subsequent literature, it argues that integrating these dimensions is crucial for sustainable urban development. ICT plays a key enabling role but must be complemented by human and social capital. Through institutional examples, such as the creation of dedicated municipal offices for digital innovation, the chapter illustrates how governance and internal capacity shape smart transitions. A human-centric approach is also essential, ensuring inclusivity, creativity, and active civic participation. Ultimately, smart cities must be viewed as cohesive urban ecosystems where technology, people, and governance interact dynamically.
Vision sensors are becoming more important in Intelligent Transportation Systems (ITS) for traffic monitoring, management, and optimization as the number of network cameras continues to rise. However, manual object tracking and matching across multiple non-overlapping cameras pose significant challenges in city-scale urban traffic scenarios. These challenges include handling diverse vehicle attributes, occlusions, illumination variations, shadows, and varying video resolutions. To address these issues, we propose an efficient and cost-effective deep learning-based framework for Multi-Object Multi-Camera Tracking (MO-MCT). The proposed framework utilizes Mask R-CNN for object detection and employs Non-Maximum Suppression (NMS) to select target objects from overlapping detections. Transfer learning is employed for re-identification, enabling the association and generation of vehicle tracklets across multiple cameras. Moreover, we leverage appropriate loss functions and distance measures to handle occlusion, illumination, and shadow challenges. The final solution identification module performs feature extraction using ResNet-152 coupled with Deep SORT based vehicle tracking. The propo
In the ten years after the Smart City was put forward, there are still problems like unclear concept, lack of top-down design and information island. With the further development of the Internet, the brain-like architecture of the Internet is becoming clearer and clearer. As a product of combination of city buildings and the Internet, the Smart City will also have a new architecture, and the city brain thus appears. Based on the Internet Brain, this paper describes how to construct the Smart City in the form of brain-like tissue, and how to evaluate the construction level of the Smart City (City IQ) relying on the Big SNS (city neural networks) and city cloud reflex arcs.
The ninth AI City Challenge continues to advance real-world applications of computer vision and AI in transportation, industrial automation, and public safety. The 2025 edition featured four tracks and saw a 17% increase in participation, with 245 teams from 15 countries registered on the evaluation server. Public release of challenge datasets led to over 30,000 downloads to date. Track 1 focused on multi-class 3D multi-camera tracking, involving people, humanoids, autonomous mobile robots, and forklifts, using detailed calibration and 3D bounding box annotations. Track 2 tackled video question answering in traffic safety, with multi-camera incident understanding enriched by 3D gaze labels. Track 3 addressed fine-grained spatial reasoning in dynamic warehouse environments, requiring AI systems to interpret RGB-D inputs and answer spatial questions that combine perception, geometry, and language. Both Track 1 and Track 3 datasets were generated in NVIDIA Omniverse. Track 4 emphasized efficient road object detection from fisheye cameras, supporting lightweight, real-time deployment on edge devices. The evaluation framework enforced submission limits and used a partially held-out test
3D city generation is a desirable yet challenging task, since humans are more sensitive to structural distortions in urban environments. Additionally, generating 3D cities is more complex than 3D natural scenes since buildings, as objects of the same class, exhibit a wider range of appearances compared to the relatively consistent appearance of objects like trees in natural scenes. To address these challenges, we propose \textbf{CityDreamer}, a compositional generative model designed specifically for unbounded 3D cities. Our key insight is that 3D city generation should be a composition of different types of neural fields: 1) various building instances, and 2) background stuff, such as roads and green lands. Specifically, we adopt the bird's eye view scene representation and employ a volumetric render for both instance-oriented and stuff-oriented neural fields. The generative hash grid and periodic positional embedding are tailored as scene parameterization to suit the distinct characteristics of building instances and background stuff. Furthermore, we contribute a suite of CityGen Datasets, including OSM and GoogleEarth, which comprises a vast amount of real-world city imagery to
The idea of smart cities (SCs) has gained substantial attention in recent years. The SC paradigm aims to improve citizens' quality of life and protect the city's environment. As we enter the age of next-generation SCs, it is important to explore all relevant aspects of the SC paradigm. In recent years, the advancement of Information and Communication Technologies (ICT) has produced a trend of supporting daily objects with smartness, targeting to make human life easier and more comfortable. The paradigm of SCs appears as a response to the purpose of building the city of the future with advanced features. SCs still face many challenges in their implementation, but increasingly more studies regarding SCs are implemented. Nowadays, different cities are employing SC features to enhance services or the residents quality of life. This work provides readers with useful and important information about Amman Smart City.
In response to challenges posed by urbanization, David Bollier from the University of Southern California raised a new idea for city planning: a comprehensive network and applications of information technologies. IBM later echoed the idea and initiated its Smart Planet vision in 2008. After that, the smart city concept was quickly adopted by major cities throughout the world, and it has gradually evolved into a strategic choice by ambitious cities. This paper looks into the smart city trend by reviewing how the concept of smart city was proposed and what the essence of a smart city is. More specifically, the driving forces of the smart city development in China are investigated, and the key differences of smart cities between China and other countries are summarized. Finally, four big challenges to build future smart cities are discussed.
With the convergence of information and telecommunication technologies, the vision of the Smart City is fast becoming a reality. City governments in a growing number of countries are capitalizing on these advances to enhance the lives of their citizens and to increase efficiency and sustainability. In this paper, we elaborate on smartCityRA, a reference architecture for Smart City projects, which serves as the design language for creating smart cities blueprints. Such a blueprint caters for diverse stakeholders, devices, platforms, and technologies. We report on our experience in carrying out a proof-of-concept use case with a major telecommunication provider in the UAE. In doing so, we refined our multiple-view model of the initial smartCityRA reference architecture. We show that Data in smart city applications drive the entire development lifecycle and should be considered early in the development cycle. In addition, Data affects all the other views in the smartCityRA and hence the Data View needs to be at the heart of the entire smartCityRA. Realizing the Data view using a component like a Data Hub helped in creating a central integration location for disparate data from differe
The task of designing a city-state of one million residents on the planet Mars seemingly approaches the limit of perceived human ingenuity. A city-state on Mars of this capacity requires us to conform to and master the simplest aspects of survival yet also allows the opportunity to handle depths that humans on Earth have yet to imagine. Because the city-state will not be the initial structure for inhabitants of Mars, a simpler colony will be instead, it is important to note the progression from the finalized colony to the city. This paper will briefly detail aspects of the successful colony, Colony NPC, to give credence to the ignition of the city-state, City NPC. Consideration of technical, economical, and societal elements of the city are also discussed.
The eighth AI City Challenge highlighted the convergence of computer vision and artificial intelligence in areas like retail, warehouse settings, and Intelligent Traffic Systems (ITS), presenting significant research opportunities. The 2024 edition featured five tracks, attracting unprecedented interest from 726 teams in 47 countries and regions. Track 1 dealt with multi-target multi-camera (MTMC) people tracking, highlighting significant enhancements in camera count, character number, 3D annotation, and camera matrices, alongside new rules for 3D tracking and online tracking algorithm encouragement. Track 2 introduced dense video captioning for traffic safety, focusing on pedestrian accidents using multi-camera feeds to improve insights for insurance and prevention. Track 3 required teams to classify driver actions in a naturalistic driving analysis. Track 4 explored fish-eye camera analytics using the FishEye8K dataset. Track 5 focused on motorcycle helmet rule violation detection. The challenge utilized two leaderboards to showcase methods, with participants setting new benchmarks, some surpassing existing state-of-the-art achievements.
Kevin Lynch proposed a theory of the image of the city identifying five elements that make the city legible or imageable. The resulting mental map of the city was conventionally derived through some qualitative processes, relying on interactions with city residents to ask them to recall city elements from their minds. This paper proposes a process by which the image of the city can be quantitatively derived automatically using computer technology and geospatial databases of the city. This method is substantially based on and inspired by Christopher Alexander's living structure and Nikos Salingaros' structural order, as a city with the living structure or structural order tends to be legible and imageable. With the increasing availability of geographic information of urban environments at very fine scales or resolutions (for example, trajectories data about human activities), the proposal or solution described in this paper is particularly timely and relevant for urban studies and architectural design. Keywords: Mental maps, head/tail division rule, legibility, imageability, power law, scaling, and hierarchy.