Update README.md
Browse files
README.md
CHANGED
|
@@ -24,10 +24,10 @@ Each sub-dataset (scene) contains:
|
|
| 24 |
|
| 25 |
- `images/`: Rendered images under different camera views and backgrounds.
|
| 26 |
- `metadata.csv`: Instance-level annotations including object attributes and positions.
|
| 27 |
-
- `causal_graph.json`: Ground-truth causal structure (as adjacency matrix or graph).
|
| 28 |
- `view_info.json`: Camera/viewpoint metadata.
|
| 29 |
- `split.json`: Recommended train/val/test splits for benchmarking.
|
| 30 |
-
|
| 31 |
---
|
| 32 |
|
| 33 |
## 🎯 Evaluation Tasks
|
|
|
|
| 24 |
|
| 25 |
- `images/`: Rendered images under different camera views and backgrounds.
|
| 26 |
- `metadata.csv`: Instance-level annotations including object attributes and positions.
|
| 27 |
+
<!-- - `causal_graph.json`: Ground-truth causal structure (as adjacency matrix or graph).
|
| 28 |
- `view_info.json`: Camera/viewpoint metadata.
|
| 29 |
- `split.json`: Recommended train/val/test splits for benchmarking.
|
| 30 |
+
-->
|
| 31 |
---
|
| 32 |
|
| 33 |
## 🎯 Evaluation Tasks
|