Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2402.18044 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929258433085440 |
|---|---|
| author | Xu, Liangyu Lu, Wanxuan Yu, Hongfeng Yao, Fanglong Sun, Xian Fu, Kun |
| author_facet | Xu, Liangyu Lu, Wanxuan Yu, Hongfeng Yao, Fanglong Sun, Xian Fu, Kun |
| contents | Extrapolating future weather radar echoes from past observations is a complex task vital for precipitation nowcasting. The spatial morphology and temporal evolution of radar echoes exhibit a certain degree of correlation, yet they also possess independent characteristics. {Existing methods learn unified spatial and temporal representations in a highly coupled feature space, emphasizing the correlation between spatial and temporal features but neglecting the explicit modeling of their independent characteristics, which may result in mutual interference between them.} To effectively model the spatiotemporal dynamics of radar echoes, we propose a Spatial-Frequency-Temporal correlation-decoupling Transformer (SFTformer). The model leverages stacked multiple SFT-Blocks to not only mine the correlation of the spatiotemporal dynamics of echo cells but also avoid the mutual interference between the temporal modeling and the spatial morphology refinement by decoupling them. Furthermore, inspired by the practice that weather forecast experts effectively review historical echo evolution to make accurate predictions, SFTfomer incorporates a joint training paradigm for historical echo sequence reconstruction and future echo sequence prediction. Experimental results on the HKO-7 dataset and ChinaNorth-2021 dataset demonstrate the superior performance of SFTfomer in short(1h), mid(2h), and long-term(3h) precipitation nowcasting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_18044 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SFTformer: A Spatial-Frequency-Temporal Correlation-Decoupling Transformer for Radar Echo Extrapolation Xu, Liangyu Lu, Wanxuan Yu, Hongfeng Yao, Fanglong Sun, Xian Fu, Kun Computer Vision and Pattern Recognition Extrapolating future weather radar echoes from past observations is a complex task vital for precipitation nowcasting. The spatial morphology and temporal evolution of radar echoes exhibit a certain degree of correlation, yet they also possess independent characteristics. {Existing methods learn unified spatial and temporal representations in a highly coupled feature space, emphasizing the correlation between spatial and temporal features but neglecting the explicit modeling of their independent characteristics, which may result in mutual interference between them.} To effectively model the spatiotemporal dynamics of radar echoes, we propose a Spatial-Frequency-Temporal correlation-decoupling Transformer (SFTformer). The model leverages stacked multiple SFT-Blocks to not only mine the correlation of the spatiotemporal dynamics of echo cells but also avoid the mutual interference between the temporal modeling and the spatial morphology refinement by decoupling them. Furthermore, inspired by the practice that weather forecast experts effectively review historical echo evolution to make accurate predictions, SFTfomer incorporates a joint training paradigm for historical echo sequence reconstruction and future echo sequence prediction. Experimental results on the HKO-7 dataset and ChinaNorth-2021 dataset demonstrate the superior performance of SFTfomer in short(1h), mid(2h), and long-term(3h) precipitation nowcasting. |
| title | SFTformer: A Spatial-Frequency-Temporal Correlation-Decoupling Transformer for Radar Echo Extrapolation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.18044 |