DsL
commited on
Commit
·
6377c65
1
Parent(s):
0cbfe76
update readme
Browse files
README.md
CHANGED
@@ -110,7 +110,6 @@ print(ds)
|
|
110 |
```
|
111 |
|
112 |
---
|
113 |
-
|
114 |
## 📌 Overview
|
115 |
|
116 |
While recent progress in AI and computer vision has been remarkable, there remains a major gap in evaluating causal reasoning over complex visual inputs. **Causal3D** bridges this gap by providing:
|
@@ -118,17 +117,6 @@ While recent progress in AI and computer vision has been remarkable, there remai
|
|
118 |
- **19 curated 3D-scene datasets** simulating diverse real-world causal phenomena.
|
119 |
- Paired **tabular causal graphs** and **image observations** across multiple views and backgrounds.
|
120 |
- Benchmarks for evaluating models in both **structured** (tabular) and **unstructured** (image) modalities.
|
121 |
-
|
122 |
-
---
|
123 |
-
|
124 |
-
## 🧩 Dataset Structure
|
125 |
-
|
126 |
-
Each sub-dataset (scene) contains:
|
127 |
-
|
128 |
-
|
129 |
-
- `images/`: Rendered images under different camera views and backgrounds.
|
130 |
-
- `tabular.csv`: Instance-level annotations including object attributes in causal graph.
|
131 |
-
|
132 |
---
|
133 |
|
134 |
## 🎯 Evaluation Tasks
|
|
|
110 |
```
|
111 |
|
112 |
---
|
|
|
113 |
## 📌 Overview
|
114 |
|
115 |
While recent progress in AI and computer vision has been remarkable, there remains a major gap in evaluating causal reasoning over complex visual inputs. **Causal3D** bridges this gap by providing:
|
|
|
117 |
- **19 curated 3D-scene datasets** simulating diverse real-world causal phenomena.
|
118 |
- Paired **tabular causal graphs** and **image observations** across multiple views and backgrounds.
|
119 |
- Benchmarks for evaluating models in both **structured** (tabular) and **unstructured** (image) modalities.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
120 |
---
|
121 |
|
122 |
## 🎯 Evaluation Tasks
|