Research map / 3D vision / NeRF and radiance fields
Seeing Through Satellite Images at Street Views
Ming Qian, Bin Tan, Qiuyu Wang, Xianwei Zheng, Hanjiang Xiong, Gui-Song Xia, Yujun Shen, Nan Xue
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 |
|---|---|---|---|
| replace Sat2Density with Sat2Density++ | street-view synthesis · CVUSA | FID: 53.29 -> 18.64 | Improved |
| replace Sat2Density with Sat2Density++ | street-view synthesis · VIGOR-OOD | FID: 85.66 -> 40.85 | Improved |
| add illumination input (ablation) | Sat2Density++ · VIGOR-OOD | FID: 73.68 -> 40.85 | Improved |
| add non-sky opacity loss (ablation) | Sat2Density++ · VIGOR-OOD | FID: 52.30 -> 40.85 | Improved |
| add sky reconstruction loss (ablation) | Sat2Density++ · VIGOR-OOD | FID: 48.78 -> 40.85 | Improved |
| add separate sky generation branch (ablation) | Sat2Density++ · VIGOR-OOD | FID: 50.10 -> 40.85 | Improved |
| use illumination-adaptive tri-plane decoder instead of vanilla (ablation) | Sat2Density++ · VIGOR-OOD | FID: 48.81 -> 40.85 | Improved |
| add GAN loss (ablation) | Sat2Density++ · VIGOR-OOD | FID: 205.62 -> 40.85 | Improved |
| add GAN loss (ablation) | Sat2Density++ · VIGOR-OOD | PSNR: 12.96 -> 12.51 | Hurt |
| add satellite-view reconstruction and discriminator losses (ablation) | Sat2Density++ · VIGOR-OOD | FID: 38.71 -> 40.85 | Mixed |
| add satellite-view losses (ablation) | Sat2Density++ · VIGOR-OOD | satellite-view PSNR: 27.5 -> 30.0 | Improved |
| use valid sky illumination for street view instead of random (ablation) | Sat2Density++ · VIGOR-OOD | FID: 64.98 -> 40.85 | Improved |
Closest papers by meaning
- Geometry-Guided Street-View Panorama Synthesis From Satellite Imagery (TPAMI 2022 · similarity 0.90)
- Geometry-Aware Satellite-to-Ground Image Synthesis for Urban Areas (CVPR 2020 · similarity 0.90)
- Coming Down to Earth: Satellite-to-Street View Synthesis for Geo-Localization (CVPR 2021 · similarity 0.88)
- Multi-Level Neural Scene Graphs for Dynamic Urban Environments (CVPR 2024 · similarity 0.85)
- Controllable Satellite-to-Street-View Synthesis with Precise Pose Alignment and Zero-Shot Environmental Control (ICLR 2025 · similarity 0.85)
- Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis (NeurIPS 2026 · 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.
