Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -10,37 +10,38 @@ tags:
|
|
10 |
- blender
|
11 |
- synthetic
|
12 |
- vision-language
|
13 |
-
pretty_name:
|
14 |
size_categories:
|
15 |
- 1K<n<100K
|
16 |
---
|
17 |
|
18 |
-
#
|
19 |
|
20 |
-
|
21 |
|
22 |
-
*
|
23 |
-
* **元数据**: 对象名称、位置、缩放、颜色、材质标志。
|
24 |
-
* **问题**: 每张图片10个,从手工制作的模板中抽取,涵盖:
|
25 |
-
* 欧几里得/水平/垂直距离查询
|
26 |
-
* 对象宽度查询
|
27 |
-
* 相对于相机的前/后谓词
|
28 |
|
29 |
-
|
|
|
|
|
|
|
|
|
30 |
|
31 |
-
|
32 |
-
|
|
|
|
|
33 |
|---------|---------|-------------------------------------|
|
34 |
-
| image
|
35 |
-
| question| string
|
36 |
-
| answer
|
37 |
|
38 |
-
##
|
39 |
```
|
40 |
@misc{synthetic_spatialvlm_qa,
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
}
|
46 |
-
```
|
|
|
10 |
- blender
|
11 |
- synthetic
|
12 |
- vision-language
|
13 |
+
pretty_name: Synthetic Spatial Visual Language Question Answering Dataset
|
14 |
size_categories:
|
15 |
- 1K<n<100K
|
16 |
---
|
17 |
|
18 |
+
# Synthetic Spatial Visual Language Question Answering Dataset
|
19 |
|
20 |
+
Automatically generated from the Blender Scene Dataset. Each example contains an image and a question-answer pair to probe **metric** (numeric) and **relation** (true/false) spatial reasoning skills.
|
21 |
|
22 |
+
* **Images**: Rendered using Blender (1000 scenes, 5 random primitives each, random cameras and lighting).
|
|
|
|
|
|
|
|
|
|
|
23 |
|
24 |
+
* **Metadata**: Object name, position, scale, color, material flags.
|
25 |
+
* **Questions**: 10 per image, drawn from handcrafted templates, covering:
|
26 |
+
* Euclidean/horizontal/vertical distance queries
|
27 |
+
* Object width queries
|
28 |
+
* Front/back predicates relative to the camera
|
29 |
|
30 |
+
Distances are in Blender scene units, rounded to 1 cm precision.
|
31 |
+
|
32 |
+
## Fields
|
33 |
+
| Field | Type | Description |
|
34 |
|---------|---------|-------------------------------------|
|
35 |
+
| image | image | rendered scene |
|
36 |
+
| question| string | natural language query |
|
37 |
+
| answer | string | ground truth answer |
|
38 |
|
39 |
+
## Quotes
|
40 |
```
|
41 |
@misc{synthetic_spatialvlm_qa,
|
42 |
+
title = {Synthetic Spatial Vision-Language Question Answering Dataset},
|
43 |
+
author = {<your name>},
|
44 |
+
year = 2025,
|
45 |
+
url = {https://huggingface.co/datasets/Litian2002/spatialvlm_qa}
|
46 |
}
|
47 |
+
```
|