Research map / 3D vision / 3D generation
MagicCity: Geometry-Aware 3D City Generation from Satellite Imagery with Multi-View Consistency
Xingbo Yao, Xuanmin Wang, Hao Wu, Chengliang Ping, Doudou Zhang, Hui Xiong
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add dual encoder embeddings (geometry and texture) over none (ablation) | MagicCity CMD · CityVista | FID: 95.23 -> 86.09 | Improved |
| add dual encoder embeddings over none (ablation) | MagicCity CMD · CityVista | Depth Error: 0.315 -> 0.137 | Improved |
| inject embeddings by cross-attention instead of direct concatenation (ablation) | MagicCity CMD · CityVista | Depth Error: 0.189 -> 0.137 | Improved |
| inject embeddings by cross-attention instead of direct concatenation (ablation) | MagicCity CMD · CityVista | Camera Error: 0.092 -> 0.071 | Improved |
| add Inter-Frame Consistency Attention (ablation) | MagicCity CMD · CityVista | Depth Error: 0.186 -> 0.137 | Improved |
| add Inter-Frame Consistency Attention (ablation) | MagicCity CMD · CityVista | FID: 89.23 -> 86.09 | Improved |
| add Hierarchical Geometric Aware attention (ablation) | MagicCity CMD · CityVista | Depth Error: 0.159 -> 0.137 | Improved |
| add Hierarchical Geometric Aware attention (ablation) | MagicCity CMD · CityVista | FID: 87.56 -> 86.09 | Improved |
| add Consistency Evaluation (ablation) | MagicCity CMD · CityVista | Camera Error: 0.102 -> 0.071 | Improved |
| add Consistency Evaluation (ablation) | MagicCity CMD · CityVista | Depth Error: 0.192 -> 0.137 | Improved |
| add Dual Encoder over none (ablation) | MagicCity CMD · CityVista | NIQE: 6.124 -> 4.552 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- CityDreamer — CityDreamer: Compositional Generative Model of Unbounded 3D Cities (CVPR 2024) · better on 6 of 6 shared results
- DreamScene — DreamScene: 3D Gaussian-based Text-to-3D Scene Generation via Formation Pattern Sampling (ECCV 2024) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Sat2City: 3D City Generation from A Single Satellite Image with Cascaded Latent Diffusion (ICCV 2025 · similarity 0.90)
- CityDreamer: Compositional Generative Model of Unbounded 3D Cities (CVPR 2024 · similarity 0.87)
- Generative Gaussian Splatting for Unbounded 3D City Generation (CVPR 2025 · similarity 0.87)
- 3D-SceneDreamer: Text-Driven 3D-Consistent Scene Generation (CVPR 2024 · similarity 0.86)
- CymbaDiff: Structured Spatial Diffusion for Sketch-based 3D Semantic Urban Scene Generation (NeurIPS 2025 · similarity 0.85)
- MagicDrive: Street View Generation with Diverse 3D Geometry Control (ICLR 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
