he RSSCN7 dataset contains satellite images acquired from Google Earth, which is originally collected for remote sensing scene classification. We conduct image synthesis on RSSCN7 to make it capable of the image inpainting task. It has seven classes: grassland, farmland, industrial and commercial regions, river and lake, forest field, residential region, and parking lot. Each class has 400 images, so there are total 2,800 images in the RSSCN7 dataset.
Variants: RSSCN7
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Classification | CPM | Convolutional Pose Machines | 2016-01-30 |
Recent papers with results on this dataset: