Research map / 3D vision / 3D generation
Leveraging BEV Paradigm for Ground-to-Aerial Image Synthesis
Junyan Ye, Jun He, Weijia Li, Zhutao Lv, Yi Lin, Jinhua Yu, Haote Yang, Conghui He
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 standard BEV transformation to street-view input (ablation) | SkyDiffusion · CVACT | FID: 62.21 -> 42.84 | Improved |
| replace standard BEV with Curved-BEV (ablation) | SkyDiffusion · CVACT | FID: 42.84 -> 36.48 | Improved |
| replace standard BEV with Curved-BEV (ablation) | SkyDiffusion · CVACT | PSNR: 13.62 -> 12.85 | Hurt |
| add standard BEV transformation (ablation) | SkyDiffusion · VIGOR-Chicago | FID: 53.27 -> 48.63 | Improved |
| add standard BEV transformation (ablation) | SkyDiffusion · VIGOR-Chicago | LPIPS: 0.666 -> 0.687 | Hurt |
| replace standard BEV with Curved-BEV (ablation) | SkyDiffusion · VIGOR-Chicago | FID: 48.63 -> 45.29 | Improved |
| add multi-to-one BEV mapping with multiple street views (ablation) | SkyDiffusion · VIGOR-Chicago | FID: 45.29 -> 31.90 | Improved |
| add multi-to-one BEV mapping (ablation) | SkyDiffusion · VIGOR-Chicago | SSIM: 0.186 -> 0.205 | Improved |
| add multi-to-one BEV mapping (ablation) | SkyDiffusion · VIGOR-Chicago | PSNR: 11.69 -> 11.15 | Hurt |
| use SkyDiffusion instead of ControlNet | SkyDiffusion · G2A-3 disaster | FID: 138.11 -> 96.16 | Improved |
| use SkyDiffusion instead of ControlNet | SkyDiffusion · G2A-3 UAV | FID: 52.76 -> 32.53 | Improved |
| use SkyDiffusion instead of ControlNet | SkyDiffusion · G2A-3 UAV | LPIPS: 0.598 -> 0.390 | Improved |
Closest papers by meaning
- AerialMegaDepth: Learning Aerial-Ground Reconstruction and View Synthesis (CVPR 2025 · similarity 0.87)
- Horizon-GS: Unified 3D Gaussian Splatting for Large-Scale Aerial-to-Ground Scenes (CVPR 2025 · similarity 0.86)
- Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis (NeurIPS 2026 · similarity 0.85)
- Satellite to GroundScape - Large-scale Consistent Ground View Generation from Satellite Views (CVPR 2025 · similarity 0.85)
- Seeing Across Skies and Streets: Feedforward 3D Reconstruction from Satellite, Drone, and Ground Images (NeurIPS 2026 · similarity 0.84)
- Controllable Satellite-to-Street-View Synthesis with Precise Pose Alignment and Zero-Shot Environmental Control (ICLR 2025 · similarity 0.84)
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.
