Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -30,10 +30,10 @@ This test set enables rigorous evaluation of object-level image captioning model
|
|
| 30 |
## 📁 Contents
|
| 31 |
|
| 32 |
- `images/`:
|
| 33 |
-
-
|
| 34 |
-
-
|
| 35 |
-
-
|
| 36 |
-
-
|
| 37 |
|
| 38 |
- `annotations.csv`: File containing metadata and ground-truth captions with columns:
|
| 39 |
- `filename`: Image file name
|
|
|
|
| 30 |
## 📁 Contents
|
| 31 |
|
| 32 |
- `images/`:
|
| 33 |
+
- `gibson/`:
|
| 34 |
+
- `gibson_rg/`: File containing the images collected using random goal exploration policy (.npy)
|
| 35 |
+
- `gibson_fr/`: File containing the images collected using frontier exploration policy (.npy)
|
| 36 |
+
- `gibson_cla/`: File containing the images collected using CLA exploration policy (.npy)
|
| 37 |
|
| 38 |
- `annotations.csv`: File containing metadata and ground-truth captions with columns:
|
| 39 |
- `filename`: Image file name
|