3D City
3D city modeling focuses on creating realistic, detailed virtual representations of urban environments for applications ranging from urban planning and autonomous driving simulation to gaming and AI training. Current research emphasizes the development of generative AI models, often employing diffusion models, transformers, and generative adversarial networks (GANs), to automatically generate city layouts and building structures from various data sources like OpenStreetMap (OSM) data, satellite imagery, and point clouds. These advancements are improving the realism, scalability, and controllability of 3D city generation, leading to more accurate and useful simulations for diverse applications.
Papers
CityGPT: Empowering Urban Spatial Cognition of Large Language Models
Jie Feng, Yuwei Du, Tianhui Liu, Siqi Guo, Yuming Lin, Yong Li
CityBench: Evaluating the Capabilities of Large Language Model as World Model
Jie Feng, Jun Zhang, Junbo Yan, Xin Zhang, Tianjian Ouyang, Tianhui Liu, Yuwei Du, Siqi Guo, Yong Li