Hugging Face's logo

Datasets:
ed1son
/
ScanBot

Tasks:
Robotics
Modalities:
Image
Text
Formats:
imagefolder
Languages:
English
Size:
< 1K
ArXiv:
Tags:
robotics
instruction-following
vision-language
laser-scanning
multimodal
Libraries:
Datasets
License:
Dataset card Data Studio Files Community
2

Add paper link and correct license

#2
by nielsr HF Staff - opened
base: refs/heads/main
←
from: refs/pr/2
Discussion Files changed
+94
-97
Files changed (1) hide show
  1. README.md +94 -97
README.md CHANGED
@@ -1,97 +1,94 @@
1
- ---
2
- pretty_name: ScanBot
3
- license: cc-by-4.0
4
- language:
5
- - en
6
- tags:
7
- - robotics
8
- - instruction-following
9
- - vision-language
10
- - laser-scanning
11
- - multimodal
12
- task_categories:
13
- - robotics
14
- dataset_info:
15
- features:
16
- - name: wrist_camera
17
- dtype: image
18
- - name: rgb_gopro
19
- dtype: image
20
- - name: instruction
21
- dtype: string
22
- - name: tcp_pose
23
- dtype: string
24
- - name: joint_states
25
- dtype: string
26
- # data_files:
27
- # - path: train/train.jsonl
28
- ---
29
-
30
-
31
- [![Project Page](https://img.shields.io/badge/Project%20Page-ScanBot-blue.svg)](https://ed1sonchen.github.io/ScanBot/)
32
-
33
-
34
- # ScanBot: Towards Intelligent Surface Scanning in Embodied Robotic Systems
35
-
36
- <img src="Figures/introduction.png" alt="ScanBot" width="100%"/>
37
-
38
-
39
- ## 🧠 Dataset Summary
40
- **ScanBot** is a dataset for instruction-conditioned, high-precision surface scanning with robots. Unlike existing datasets that focus on coarse tasks like grasping or navigation, ScanBot targets industrial laser scanning, where sub-millimeter accuracy and parameter stability are essential. It includes scanning trajectories across 12 objects and 6 task types, each driven by natural language instructions. The dataset provides synchronized RGB, depth, laser profiles, robot poses, and joint states.
41
-
42
-
43
- ## 📦 Use Cases
44
-
45
- - Vision-Language Action (VLA)
46
- - Instruction-Guided Surface Scanning
47
- - 3D Surface Reconstruction
48
- - Spatial Reasoning and Feature Localization
49
- - Laser Profile Analysis for Inspection Tasks
50
-
51
-
52
-
53
- ## 🗂️ Data Description
54
- ```
55
- scanbot/
56
- ├── cube1/
57
- │ ├── top_surface/
58
- │ │ ├── path_001/
59
- │ │ │ ├── rgb/
60
- │ │ │ │ └── ...
61
- │ │ │ ├── depth/
62
- │ │ │ │ └── ...
63
- | | | |── 1746226187.997976_gopro.mp4
64
- │ │ │ ├── robot_joint_states.csv
65
- │ │ │ ├── robot_tcp_poses.csv
66
- │ │ │ └── metadata.json
67
- │ │ ├── path_002/
68
- │ │ │ └── ...
69
- ├── cube2
70
- ├── cylinder_red
71
- ├── cylinder_white
72
- └── ...
73
- ```
74
-
75
- ## 💬 Task Levels
76
-
77
- Each scan is driven by an instruction that falls under one of the following task categories:
78
-
79
- | Task Type | Description |
80
- |-----------|-------------|
81
- | T1 | Surface Scan |
82
- | T2 | Geometry Focus |
83
- | T3 | Spatial Reference |
84
- | T4 | Functional Target |
85
- | T5 | Defect Inspection |
86
- | T6 | Comparative Analysis |
87
-
88
- ## 🛠️ Hardware Setup
89
-
90
- <img src="Figures/setup.png" alt="ScanBot Example" width="30%"/>
91
-
92
- ## 📖 Version Update
93
-
94
- ### Version 1.0
95
- The initial version of ScanBot contains 197 tasks, and involves 12 different objects.
96
-
97
-
 
1
+ ---
2
+ language:
3
+ - en
4
+ license: apache-2.0
5
+ task_categories:
6
+ - robotics
7
+ pretty_name: ScanBot
8
+ tags:
9
+ - robotics
10
+ - instruction-following
11
+ - vision-language
12
+ - laser-scanning
13
+ - multimodal
14
+ dataset_info:
15
+ features:
16
+ - name: wrist_camera
17
+ dtype: image
18
+ - name: rgb_gopro
19
+ dtype: image
20
+ - name: instruction
21
+ dtype: string
22
+ - name: tcp_pose
23
+ dtype: string
24
+ - name: joint_states
25
+ dtype: string
26
+ ---
27
+
28
+ [![Project Page](https://img.shields.io/badge/Project%20Page-ScanBot-blue.svg)](https://ed1sonchen.github.io/ScanBot/)
29
+
30
+
31
+ # ScanBot: Towards Intelligent Surface Scanning in Embodied Robotic Systems
32
+
33
+ <img src="Figures/introduction.png" alt="ScanBot" width="100%"/>
34
+
35
+ This dataset is presented in the paper [ScanBot: Towards Intelligent Surface Scanning in Embodied Robotic Systems](https://huggingface.co/papers/2505.17295).
36
+
37
+
38
+ ## 🧠 Dataset Summary
39
+ **ScanBot** is a dataset for instruction-conditioned, high-precision surface scanning with robots. Unlike existing datasets that focus on coarse tasks like grasping or navigation, ScanBot targets industrial laser scanning, where sub-millimeter accuracy and parameter stability are essential. It includes scanning trajectories across 12 objects and 6 task types, each driven by natural language instructions. The dataset provides synchronized RGB, depth, laser profiles, robot poses, and joint states.
40
+
41
+
42
+ ## 📦 Use Cases
43
+
44
+ - Vision-Language Action (VLA)
45
+ - Instruction-Guided Surface Scanning
46
+ - 3D Surface Reconstruction
47
+ - Spatial Reasoning and Feature Localization
48
+ - Laser Profile Analysis for Inspection Tasks
49
+
50
+
51
+
52
+ ## 🗂️ Data Description
53
+ ```
54
+ scanbot/
55
+ ├── cube1/
56
+ │ ├── top_surface/
57
+ │ │ ├── path_001/
58
+ │ │ │ ├── rgb/
59
+ │ │ │ │ └── ...
60
+ │ │ │ ├── depth/
61
+ │ │ │ │ └── ...
62
+ | | | |── 1746226187.997976_gopro.mp4
63
+ │ │ │ ├── robot_joint_states.csv
64
+ │ │ │ ├── robot_tcp_poses.csv
65
+ │ │ │ └── metadata.json
66
+ │ │ ├── path_002/
67
+ │ │ │ └── ...
68
+ ├── cube2
69
+ ├── cylinder_red
70
+ ├── cylinder_white
71
+ └── ...
72
+ ```
73
+
74
+ ## 💬 Task Levels
75
+
76
+ Each scan is driven by an instruction that falls under one of the following task categories:
77
+
78
+ | Task Type | Description |
79
+ |-----------|-------------|
80
+ | T1 | Surface Scan |
81
+ | T2 | Geometry Focus |
82
+ | T3 | Spatial Reference |
83
+ | T4 | Functional Target |
84
+ | T5 | Defect Inspection |
85
+ | T6 | Comparative Analysis |
86
+
87
+ ## 🛠️ Hardware Setup
88
+
89
+ <img src="Figures/setup.png" alt="ScanBot Example" width="30%"/>
90
+
91
+ ## 📖 Version Update
92
+
93
+ ### Version 1.0
94
+ The initial version of ScanBot contains 197 tasks, and involves 12 different objects.