TiMo: Spatiotemporal Foundation Model for Satellite Image Time Series
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918019173711872 |
|---|---|
| author | Qin, Xiaolei Wang, Di Zhang, Jing Wang, Fengxiang Su, Xin Du, Bo Zhang, Liangpei |
| author_facet | Qin, Xiaolei Wang, Di Zhang, Jing Wang, Fengxiang Su, Xin Du, Bo Zhang, Liangpei |
| contents | Satellite image time series (SITS) provide continuous observations of the Earth's surface, making them essential for applications such as environmental management and disaster assessment. However, existing spatiotemporal foundation models rely on plain vision transformers, which encode entire temporal sequences without explicitly capturing multiscale spatiotemporal relationships between land objects. This limitation hinders their effectiveness in downstream tasks. To overcome this challenge, we propose TiMo, a novel hierarchical vision transformer foundation model tailored for SITS analysis. At its core, we introduce a spatiotemporal gyroscope attention mechanism that dynamically captures evolving multiscale patterns across both time and space. For pre-training, we curate MillionST, a large-scale dataset of one million images from 100,000 geographic locations, each captured across 10 temporal phases over five years, encompassing diverse geospatial changes and seasonal variations. Leveraging this dataset, we adapt masked image modeling to pre-train TiMo, enabling it to effectively learn and encode generalizable spatiotemporal representations.Extensive experiments across multiple spatiotemporal tasks-including deforestation monitoring, land cover segmentation, crop type classification, and flood detection-demonstrate TiMo's superiority over state-of-the-art methods. Code, model, and dataset will be released at https://github.com/MiliLab/TiMo. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_08723 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TiMo: Spatiotemporal Foundation Model for Satellite Image Time Series Qin, Xiaolei Wang, Di Zhang, Jing Wang, Fengxiang Su, Xin Du, Bo Zhang, Liangpei Computer Vision and Pattern Recognition Satellite image time series (SITS) provide continuous observations of the Earth's surface, making them essential for applications such as environmental management and disaster assessment. However, existing spatiotemporal foundation models rely on plain vision transformers, which encode entire temporal sequences without explicitly capturing multiscale spatiotemporal relationships between land objects. This limitation hinders their effectiveness in downstream tasks. To overcome this challenge, we propose TiMo, a novel hierarchical vision transformer foundation model tailored for SITS analysis. At its core, we introduce a spatiotemporal gyroscope attention mechanism that dynamically captures evolving multiscale patterns across both time and space. For pre-training, we curate MillionST, a large-scale dataset of one million images from 100,000 geographic locations, each captured across 10 temporal phases over five years, encompassing diverse geospatial changes and seasonal variations. Leveraging this dataset, we adapt masked image modeling to pre-train TiMo, enabling it to effectively learn and encode generalizable spatiotemporal representations.Extensive experiments across multiple spatiotemporal tasks-including deforestation monitoring, land cover segmentation, crop type classification, and flood detection-demonstrate TiMo's superiority over state-of-the-art methods. Code, model, and dataset will be released at https://github.com/MiliLab/TiMo. |
| title | TiMo: Spatiotemporal Foundation Model for Satellite Image Time Series |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.08723 |