RainUNet for Super-Resolution Rain Movie Prediction under Spatio-temporal Shifts

TL;DR

This paper introduces RainUNet, a hierarchical U-shaped neural network with temporal-wise separable blocks, designed to forecast high-resolution rainfall movies from satellite data, effectively addressing spatio-temporal shifts in weather prediction.

Contribution

The paper proposes RainUNet, a novel neural network architecture with decoupled large kernel 3D convolutions, tailored for high-resolution rainfall prediction from satellite imagery.

Findings

RainUNet outperforms baseline methods in various evaluation metrics.

The hierarchical structure and TS blocks improve spatio-temporal modeling.

Source code is publicly available for reproducibility.

Abstract

This paper presents a solution to the Weather4cast 2022 Challenge Stage 2. The goal of the challenge is to forecast future high-resolution rainfall events obtained from ground radar using low-resolution multiband satellite images. We suggest a solution that performs data preprocessing appropriate to the challenge and then predicts rainfall movies using a novel RainUNet. RainUNet is a hierarchical U-shaped network with temporal-wise separable block (TS block) using a decoupled large kernel 3D convolution to improve the prediction performance. Various evaluation metrics show that our solution is effective compared to the baseline method. The source codes are available at https://github.com/jinyxp/Weather4cast-2022