Research map / 3D vision / Gaussian splatting
station2radar: query‑conditioned gaussian splatting for precipitation field
Doyi Kim, Minseok Seo, Changick Kim
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 AWS fusion to U-Net (ablation) | satellite-to-radar U-Net (ConvNeXt) · snapshot | CSI: 0.62 -> 0.73 | Improved |
| add Gaussian splatting rendering to AWS-fused U-Net (ablation) | QCGS · snapshot | CSI: 0.73 -> 0.76 | Improved |
| fuse AWS only without satellite (ablation) | AWS-only variant · snapshot | CSI: 0.76 -> 0.53 | Hurt |
| combine gradient, regular and heavy-rain sampling vs regular only (ablation) | QCGS query sampling · snapshot | CSI: 0.68 -> 0.76 | Improved |
| add heavy-rain sampling to regular+gradient (ablation) | QCGS query sampling · snapshot | CSI: 0.73 -> 0.76 | Improved |
| add regular sampling to gradient+heavy (ablation) | QCGS query sampling · snapshot | CSI: 0.74 -> 0.76 | Improved |
| increase query points K from 1000 to 6000 (ablation) | QCGS · snapshot | CSI: 0.69 -> 0.76 | Improved |
| increase query points K from 6000 to 9000 (ablation) | QCGS · snapshot | CSI: 0.76 -> 0.77 | No clear effect |
| QCGS trained at 2 km vs data-driven BBDM at 0.5 km | QCGS · snapshot | RMSE: 1.68 -> 1.00 | Improved |
| QCGS vs Kriging interpolation | QCGS · snapshot | CSI: 0.50 -> 0.76 | Improved |
| QCGS vs IMERG on daily accumulation | QCGS · daily accumulation | RMSE: 14.08 -> 6.68 | Improved |
| QCGS vs MSWEP on daily accumulation CSI | QCGS · daily accumulation | CSI: 0.95 -> 0.93 | Hurt |
Closest papers by meaning
- PrecipDiff: Leveraging Image Diffusion Models to Enhance Satellite-Based Precipitation Observations (AAAI 2025 · similarity 0.86)
- Data-driven Precipitation Nowcasting Using Satellite Imagery (AAAI 2025 · similarity 0.86)
- High-Dynamic Radar Sequence Prediction for Weather Nowcasting Using Spatiotemporal Coherent Gaussian Representation (ICLR 2025 · similarity 0.86)
- AlphaPre: Amplitude-Phase Disentanglement Model for Precipitation Nowcasting (CVPR 2025 · similarity 0.85)
- PostCast: Generalizable Postprocessing for Precipitation Nowcasting via Unsupervised Blurriness Modeling (ICLR 2025 · similarity 0.84)
- Precipitation Downscaling with Spatiotemporal Video Diffusion (NeurIPS 2024 · 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.
