DINOv3 as a Frozen Encoder for CRPS-Oriented Probabilistic Rainfall Nowcasting
Not on the map. Why?
Abstract (arXiv)
This paper proposes a competitive and computationally efficient approach to probabilistic rainfall nowcasting. A video projector (V-JEPA Vision Transformer) associated to a lightweight probabilistic head is attached to a pre-trained satellite vision encoder (DINOv3-SAT493M) to map encoder tokens into a discrete empirical CDF (eCDF) over 4-hour accumulated rainfall. The projector-head is optimized end-to-end over the Ranked Probability Score (RPS). As an alternative, 3D-UNET baselines trained with an aggregate Rank Probability Score and a per-pixel Gamma-Hurdle objective are used. On the Weather4Cast 2025 benchmark, the proposed method achieved a promising performance, with a CRPS of 3.5102, which represents ≈ 26% in effectiveness gain against the best 3D-UNET.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 9 of its 18 references that are in the map. First posted on arXiv 14 Nov 2025.
Builds on
9 of 18 references.
References · 9
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
- 20163D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotationcited by 1,476
- 2015Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcastingcited by 1,436
- 2017Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semanticscited by 1,358
- 2021Skillful Precipitation Nowcasting using Deep Generative Models of Radarcited by 311
- 2020MetNet: A Neural Weather Model for Precipitation Forecastingcited by 131
- 2022Mitigating Negative Transfer in Multi-Task Learning with Exponential Moving Average Loss Weighting Strategiescited by 10
- 2022Super-resolution Probabilistic Rain Prediction from Satellite Data Using 3D U-Nets and EarthFormerscited by 5
- 2022Solving the Weather4cast Challenge via Visual Transformers for 3D Imagescited by 2
Built on by
Not cited yet.