Research map / Image generation / Autoregressive and latent generation
SEVIR : A Storm Event Imagery Dataset for Deep Learning Applications in Radar and Satellite Meteorology
Mark Veillette, Siddharth Samsi, Chris Mattioli
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 VGG content loss to MSE (ablation) | U-Net synthetic radar · SEVIR synthetic radar | CSI at thresholds 74 and 133: 0.4273 -> 0.4378 (thr 74); 0.2162 -> 0.2263 (thr 133) | Improved |
| add VGG content loss to MSE (ablation) | U-Net synthetic radar · SEVIR synthetic radar | MSE: 466.64 -> 497.26 | Hurt |
| replace MSE with cGAN+MAE (ablation) | U-Net synthetic radar · SEVIR synthetic radar | MAE: 10.93 -> 12.69 | Hurt |
| replace MSE with cGAN+MAE (ablation) | U-Net synthetic radar · SEVIR synthetic radar | CSI thr 74: 0.4273 -> 0.2803 | Hurt |
| replace MSE with cGAN+MAE (ablation) | U-Net synthetic radar · SEVIR synthetic radar | LPIPS | Improved |
| add style/content (SC) loss to MSE (ablation) | U-Net nowcast · SEVIR nowcasting | texture and motion | Improved |
| train with SC loss only (no reconstruction loss) (ablation) | U-Net nowcast · SEVIR nowcasting | forecast motion/CSI | Hurt |
| use cGAN+MAE instead of MSE (ablation) | U-Net nowcast · SEVIR nowcasting | CSI threshold 74 at leads >25 min | Improved |
| use U-Net MSE/MSE+SC instead of optical flow (ablation) | U-Net nowcast · SEVIR nowcasting | CSI | Hurt |
Closest papers by meaning
- HR-Extreme: A High-Resolution Dataset for Extreme Weather Forecasting (ICLR 2025 · similarity 0.83)
- Scaling Storm-Resolving Atmospheric AI Simulation to the Entire Planet (NeurIPS 2026 · similarity 0.83)
- Data-driven Precipitation Nowcasting Using Satellite Imagery (AAAI 2025 · similarity 0.83)
- PrecipDiff: Leveraging Image Diffusion Models to Enhance Satellite-Based Precipitation Observations (AAAI 2025 · similarity 0.83)
- MFogHub: Bridging Multi-Regional and Multi-Satellite Data for Global Marine Fog Detection and Forecasting (CVPR 2025 · similarity 0.82)
- AlphaPre: Amplitude-Phase Disentanglement Model for Precipitation Nowcasting (CVPR 2025 · similarity 0.82)
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.
