Research map / 3D vision / Gaussian splatting
Cross-View Splatter: Feed-Forward View Synthesis with Georeferenced Images
Matias Turkulainen, Akshay Krishnan, Filippo Aleotti, Mohamed Sayed, Guillermo Garcia-Hernando, Juho Kannala, Arno Solin, Gabriel Brostow, Daniyar Turmukhambetov
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 satellite terrain Gaussians to ground Gaussians (ablation) | Cross-View Splatter · Tanks and Temples outdoor (10 scenes), 1 context view | PSNR: 8.92 -> 11.13 | Improved |
| add satellite terrain Gaussians to ground Gaussians (ablation) | Cross-View Splatter · Tanks and Temples outdoor (10 scenes), 2 context views | PSNR: 9.94 -> 11.67 | Improved |
| add satellite terrain Gaussians to ground Gaussians (ablation) | Cross-View Splatter · Tanks and Temples outdoor (10 scenes), 3 context views | PSNR: 10.61 -> 12.00 | Improved |
| add satellite terrain Gaussians to ground Gaussians (ablation) | Cross-View Splatter · DL3DV outdoor (40 scenes), 1 context view | PSNR: 9.00 -> 11.33 | Improved |
| add satellite terrain Gaussians to ground Gaussians (ablation) | Cross-View Splatter · DL3DV outdoor (40 scenes), 3 context views | PSNR: 10.65 -> 12.61 | Improved |
| add ground Gaussians to terrain-only prediction (ablation) | Cross-View Splatter · Tanks and Temples outdoor (10 scenes), 1 context view | PSNR: 8.39 -> 11.13 | Improved |
| fine-tune from AnySplat with satellite conditioning (vs AnySplat) | Cross-View Splatter · Tanks and Temples outdoor (10 scenes), 1 context view | PSNR: 7.48 -> 11.13 | Improved |
| fine-tune from AnySplat with satellite conditioning (vs AnySplat) | Cross-View Splatter · DL3DV outdoor (40 scenes), 1 context view | PSNR: 8.37 -> 11.33 | Improved |
| add GPS translation noise up to 5m (ablation) | Cross-View Splatter (Combined) · Tanks and Temples outdoor (10 scenes), 1 context view | PSNR: 11.13 -> 11.12 ± 0.04 | No clear effect |
| add GPS rotation noise up to 15 degrees (ablation) | Cross-View Splatter (Combined) · Tanks and Temples outdoor (10 scenes), 1 context view | PSNR: 11.13 -> 11.11 ± 0.12 | No clear effect |
| compare with diffusion model SEVA at 2 context views | Cross-View Splatter (Combined) · Tanks and Temples outdoor (10 scenes), 2 context views | PSNR: 12.09 -> 11.67 | Hurt |
| compare with diffusion model SEVA at 1 context view | Cross-View Splatter (Combined) · Tanks and Temples outdoor (10 scenes), 1 context view | PSNR: 10.39 -> 11.13 | Improved |
Closest papers by meaning
- Horizon-GS: Unified 3D Gaussian Splatting for Large-Scale Aerial-to-Ground Scenes (CVPR 2025 · similarity 0.90)
- RPC-GS: Gaussian Splatting with native RPC Rendering for Satellite Imagery (NeurIPS 2026 · similarity 0.89)
- SplatFormer: Point Transformer for Robust 3D Gaussian Splatting (ICLR 2025 · similarity 0.88)
- Splatter Image: Ultra-Fast Single-View 3D Reconstruction (CVPR 2024 · similarity 0.88)
- Satellite to GroundScape - Large-scale Consistent Ground View Generation from Satellite Views (CVPR 2025 · similarity 0.87)
- Wild-GS: Real-Time Novel View Synthesis from Unconstrained Photo Collections (NeurIPS 2024 · similarity 0.87)
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.
