The xBD dataset contains over 45,000KM2 of polygon labeled pre and post disaster imagery. The dataset provides the post-disaster imagery with transposed polygons from pre over the buildings, with damage classification labels.
Source: xBD
Image Source: Gupta et al
Variants: xBD
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| 2D Semantic Segmentation | MambaBDA-Base | ChangeMamba: Remote Sensing Change Detection … | 2024-04-04 |
| 2D Semantic Segmentation | BDANet | BDANet: Multiscale Convolutional Neural Network … | 2021-05-16 |
| 2D Semantic Segmentation | Double branch U-Net | Cross-directional Feature Fusion Network for … | 2020-10-27 |
| 2D Semantic Segmentation | Double branch U-Net | Building Disaster Damage Assessment in … | 2020-04-12 |
| 2D Semantic Segmentation | Baseline Model | xBD: A Dataset for Assessing … | 2019-11-21 |
Recent papers with results on this dataset: