Foundation Models for Generalist Geospatial Artificial Intelligence
- 论文名称: Foundation Models for Generalist Geospatial Artificial Intelligence
- Link: https://arxiv.org/abs/2310.18660
- Published in: Arxiv 2023
- Type: Pretrain
- Code/Project: https://huggingface.co/ibm-nasa-geospatial
- 备注: MAE, HLS多光谱
- 数据类型: 多光谱
- Backbone: ViT
- 下游任务: Cloud Gap Imputation/Flood Mapping/Wildfire Scar Mapping/Crop Segmentation
- Short Summary: 对ViT架构的主要修改是3D位置嵌入和3D补丁嵌入,以处理时空数据
This post is licensed under CC BY 4.0 by the author.