Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -31,6 +31,56 @@ The creation of this dataset involved three key activities:
|
|
| 31 |
|
| 32 |
This dataset was introduced in our **paper/repository**: [RSDiX: Lightweight and Data-Efficient VLMs for Remote Sensing through Self-Distillation](https://github.com/NeuRoNeLab/RSDiX-CLIP). Please refer to it for further details on data collection, captioning methodology, and use cases.
|
| 33 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
## Dataset Structure
|
| 35 |
The `dataset_s2lcd.json` file contains caption annotations for Sentinel-2 image patches. Each entry in the `images` array corresponds to a single image patch and includes:
|
| 36 |
|
|
|
|
| 31 |
|
| 32 |
This dataset was introduced in our **paper/repository**: [RSDiX: Lightweight and Data-Efficient VLMs for Remote Sensing through Self-Distillation](https://github.com/NeuRoNeLab/RSDiX-CLIP). Please refer to it for further details on data collection, captioning methodology, and use cases.
|
| 33 |
|
| 34 |
+
|
| 35 |
+
<table>
|
| 36 |
+
<tr>
|
| 37 |
+
<td align="center">
|
| 38 |
+
<img src="S2LCD_images/S2A_L1C_20150925_N0204R065_160.tif" width="200"><br>
|
| 39 |
+
<strong>References:</strong><br>
|
| 40 |
+
<em>
|
| 41 |
+
Coastal hamlet near forest and water body.<br>
|
| 42 |
+
Water body surrounded by buildings and forest.<br>
|
| 43 |
+
Small coastal built-up area.<br>
|
| 44 |
+
Dense forestry coverage and water body.<br>
|
| 45 |
+
Abiotic artificial surface and forest coverage.
|
| 46 |
+
</em>
|
| 47 |
+
</td>
|
| 48 |
+
<td align="center">
|
| 49 |
+
<img src="S2LCD_images/S2A_L1C_20150925_N0204R065_751.tif" width="200"><br>
|
| 50 |
+
<strong>References:</strong><br>
|
| 51 |
+
<em>
|
| 52 |
+
Rocky areas with spread vegetation and sporadic snow coverage.<br>
|
| 53 |
+
Abiotic natural surface with permanent herbaceous and forest areas near snow coverage.<br>
|
| 54 |
+
Dense forest and bare soil.<br>
|
| 55 |
+
Wood and herbaceous coverage in mountain context.<br>
|
| 56 |
+
Forest area next to grasslands and snow coverage.
|
| 57 |
+
</em>
|
| 58 |
+
</td>
|
| 59 |
+
<td align="center">
|
| 60 |
+
<img src="S2LCD_images/S2A_L1C_20150925_N0204R065_363.tif" width="200"><br>
|
| 61 |
+
<strong>References:</strong><br>
|
| 62 |
+
<em>
|
| 63 |
+
Lake surrounded by herbaceous coverage and sporadic trees.<br>
|
| 64 |
+
Permanent herbaceous coverage surrounding a water body, also close to artificial abiotic surfaces.<br>
|
| 65 |
+
Woody areas among the grasslands near the town.<br>
|
| 66 |
+
Small areas of agricultural fields between grasslands with trees in line.<br>
|
| 67 |
+
Residential area with trees in line near grasslands and woody areas.
|
| 68 |
+
</em>
|
| 69 |
+
</td>
|
| 70 |
+
<td align="center">
|
| 71 |
+
<img src="S2LCD_images/S2A_L1C_20150925_N0204R065_2381.tif" width="200"><br>
|
| 72 |
+
<strong>References:</strong><br>
|
| 73 |
+
<em>
|
| 74 |
+
Airport areas and a small water body near extensive and intensive cultivated fields <br> with the presence of scattered and isolated buildings and a low density hamlet.<br>
|
| 75 |
+
Isolated built-up surrounded by cultivated fields near an airport.<br>
|
| 76 |
+
Small water body next to an airport.<br>
|
| 77 |
+
Scattered buildings and a small hamlet surrounded by cultivated fields.
|
| 78 |
+
</em>
|
| 79 |
+
</td>
|
| 80 |
+
</tr>
|
| 81 |
+
</table>
|
| 82 |
+
|
| 83 |
+
|
| 84 |
## Dataset Structure
|
| 85 |
The `dataset_s2lcd.json` file contains caption annotations for Sentinel-2 image patches. Each entry in the `images` array corresponds to a single image patch and includes:
|
| 86 |
|