PTCT: Patches with 3D-Temporal Convolutional Transformer Network for Precipitation Nowcasting
This work addresses precipitation nowcasting for meteorology and disaster management, offering an incremental improvement over existing methods by combining patches and convolution in a Transformer framework.
The paper tackles precipitation nowcasting by predicting future rainfall intensity from radar echo sequences, proposing PTCT, a Transformer variant that splits frames into patches and uses 3D-temporal convolution to capture short-term dependencies, achieving state-of-the-art performance on two radar echo datasets.
Precipitation nowcasting is to predict the future rainfall intensity over a short period of time, which mainly relies on the prediction of radar echo sequences. Though convolutional neural network (CNN) and recurrent neural network (RNN) are widely used to generate radar echo frames, they suffer from inductive bias (i.e., translation invariance and locality) and seriality, respectively. Recently, Transformer-based methods also gain much attention due to the great potential of Transformer structure, whereas short-term dependencies and autoregressive characteristic are ignored. In this paper, we propose a variant of Transformer named patches with 3D-temporal convolutional Transformer network (PTCT), where original frames are split into multiple patches to remove the constraint of inductive bias and 3D-temporal convolution is employed to capture short-term dependencies efficiently. After training, the inference of PTCT is performed in an autoregressive way to ensure the quality of generated radar echo frames. To validate our algorithm, we conduct experiments on two radar echo dataset: Radar Echo Guangzhou and HKO-7. The experimental results show that PTCT achieves state-of-the-art (SOTA) performance compared with existing methods.