Update README.md
Browse files
README.md
CHANGED
|
@@ -21,8 +21,13 @@ While recent progress in AI and computer vision has been remarkable, there remai
|
|
| 21 |
|
| 22 |
Each sub-dataset (scene) contains:
|
| 23 |
|
|
|
|
| 24 |
- `images/`: Rendered images under different camera views and backgrounds.
|
| 25 |
- `tabular.csv`: Instance-level annotations including object attributes in causal graph.
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
<!-- - `causal_graph.json`: Ground-truth causal structure (as adjacency matrix or graph).
|
| 27 |
- `view_info.json`: Camera/viewpoint metadata.
|
| 28 |
- `split.json`: Recommended train/val/test splits for benchmarking. -->
|
|
@@ -55,6 +60,8 @@ We evaluate a diverse set of methods:
|
|
| 55 |
|
| 56 |
---
|
| 57 |
|
|
|
|
|
|
|
| 58 |
<!-- ## 🔍 Example Use Case
|
| 59 |
|
| 60 |
```python
|
|
|
|
| 21 |
|
| 22 |
Each sub-dataset (scene) contains:
|
| 23 |
|
| 24 |
+
|
| 25 |
- `images/`: Rendered images under different camera views and backgrounds.
|
| 26 |
- `tabular.csv`: Instance-level annotations including object attributes in causal graph.
|
| 27 |
+
📷 **Example image** from the dataset:
|
| 28 |
+
|
| 29 |
+

|
| 30 |
+
|
| 31 |
<!-- - `causal_graph.json`: Ground-truth causal structure (as adjacency matrix or graph).
|
| 32 |
- `view_info.json`: Camera/viewpoint metadata.
|
| 33 |
- `split.json`: Recommended train/val/test splits for benchmarking. -->
|
|
|
|
| 60 |
|
| 61 |
---
|
| 62 |
|
| 63 |
+
|
| 64 |
+
|
| 65 |
<!-- ## 🔍 Example Use Case
|
| 66 |
|
| 67 |
```python
|