Sat2City v2 generates 3D city assets from satellite images

Researchers have introduced Sat2City v2, an advanced framework for generating 3D city assets from a single satellite image. This new version improves upon its predecessor by adapting a pre-trained 3D foundation model to work with weakly aligned satellite images and textured meshes. Sat2City v2 utilizes a real-world dataset of over 16,000 satellite-mesh pairs and focuses on creating explicit textured mesh assets rather than just rendering proxies, achieving superior performance on geometry and appearance benchmarks. AI

IMPACT Advances the creation of detailed 3D city models for applications like digital twins and urban simulation.

RANK_REASON The item is a research paper detailing a new version of a computer vision model for 3D asset generation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• Connected Papers
• Core Recommendations for Antifungal Stewardship: A Statement of the Mycoses Study Group Education and Research Consortium
• DagsHub
• Hugging Face
• International Conference on Computer Vision
• Litmaps
• Sat2City
• Sat2City v2

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →