When the Past Matters: FlashBack Memory for Precipitation Nowcasting
2026-06-15 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors focused on improving short-term rain forecasts by creating a new tool called FlashBack Memory (FB). FB helps models better remember important past weather data and use it wisely, which leads to fewer false alarms and missed rain events. They tested FB with several existing prediction models and found it improved accuracy, especially during heavy rain and longer forecast times. Overall, their approach makes rain predictions more reliable and precise.
precipitation nowcastingspatiotemporal resolutionrecurrent neural networksmemory moduleadaptive fusion gatePredRNNMSEMAESSIMCSI
Authors
Yuhao Du, Boxiao Huang, Chengrong Wu, Jiankai Zhang
Abstract
Accurate precipitation nowcasting is crucial for disaster mitigation and socio-economic planning, yet existing methods often struggle with false alarms, missed events, and long range dependency modeling at high spatiotemporal resolution. To address these challenges, we propose FlashBack Memory (FB), a module that dynamically retrieves key historical states and integrates them via an adaptive fusion gate, enhancing the spatiotemporal representation capability of recurrent-based models. We incorporate FB into PredRNN, PredRNNpp, MIM, MotionRNN, and PredRNN-V2, and evaluate on CIKM2017, Shanghai2020, and SEVIR datasets. Experimental results demonstrate that FB significantly improves MSE, MAE, SSIM, and CSI metrics, particularly for high-intensity rainfall and long-sequence predictions, while reducing false alarms and missed events and enhancing temporal consistency and spatial localization. The proposed method provides a general and efficient memory enhancement mechanism, improving the overall performance of recurrent-based precipitation nowcasting models.