Update README.md
Browse files
README.md
CHANGED
|
@@ -1,19 +1,4 @@
|
|
| 1 |
-
# ArticuBot Dataset (WebDataset Format)
|
| 2 |
-
|
| 3 |
-
## Dataset Statistics
|
| 4 |
-
|
| 5 |
-
- **Total Samples**: 20,963
|
| 6 |
-
- **Total Timesteps**: 2,299,385
|
| 7 |
-
- **Categories**: ['Dishwasher', 'Microwave', 'Oven', 'Refrigerator', 'StorageFurniture', 'unknown']
|
| 8 |
-
- **Splits**: ['train']
|
| 9 |
-
|
| 10 |
-
## Split Statistics
|
| 11 |
-
|
| 12 |
-
### train
|
| 13 |
-
- **Samples**: 20,963
|
| 14 |
-
- **Timesteps**: 2,299,385
|
| 15 |
-
- **Categories**: 6
|
| 16 |
-
- **Failed Samples**: 0
|
| 17 |
|
| 18 |
## Data Format
|
| 19 |
|
|
@@ -32,40 +17,15 @@ Each trajectory contains timesteps with the following data:
|
|
| 32 |
- `displacement_gripper_to_object`: Spatial displacement vectors
|
| 33 |
- `goal_gripper_pcd`: Goal gripper point cloud
|
| 34 |
|
| 35 |
-
## Usage
|
| 36 |
-
|
| 37 |
-
```python
|
| 38 |
-
import webdataset as wds
|
| 39 |
-
import pickle
|
| 40 |
-
import json
|
| 41 |
-
|
| 42 |
-
# Load training data
|
| 43 |
-
dataset = wds.WebDataset("articubot_train_*.tar")
|
| 44 |
-
dataset = dataset.decode()
|
| 45 |
-
|
| 46 |
-
for sample in dataset:
|
| 47 |
-
# Load metadata
|
| 48 |
-
metadata = json.loads(sample["metadata.json"])
|
| 49 |
-
|
| 50 |
-
# Load trajectory data
|
| 51 |
-
trajectory_data = pickle.loads(sample["trajectory.pkl"])
|
| 52 |
-
|
| 53 |
-
# Process trajectory data
|
| 54 |
-
for timestep in trajectory_data:
|
| 55 |
-
state = timestep["state"]
|
| 56 |
-
action = timestep["action"]
|
| 57 |
-
# ... process other fields
|
| 58 |
-
```
|
| 59 |
|
| 60 |
## Citation
|
| 61 |
|
| 62 |
-
If you use this dataset in your research, please cite:
|
| 63 |
|
| 64 |
```bibtex
|
| 65 |
-
@
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
}
|
| 71 |
```
|
|
|
|
| 1 |
+
# ArticuBot Simulated Dataset for Trajectories of Articulation (saved in WebDataset Format and splitter into train / val)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
|
| 3 |
## Data Format
|
| 4 |
|
|
|
|
| 17 |
- `displacement_gripper_to_object`: Spatial displacement vectors
|
| 18 |
- `goal_gripper_pcd`: Goal gripper point cloud
|
| 19 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
|
| 21 |
## Citation
|
| 22 |
|
| 23 |
+
If you use this dataset in your research, please cite original AricuBot paper:
|
| 24 |
|
| 25 |
```bibtex
|
| 26 |
+
@inproceedings{Wang2025articubot,
|
| 27 |
+
title={ArticuBot: Learning Universal Articulated Object Manipulation Policy via Large Scale Simulation},
|
| 28 |
+
author={Wang, Yufei and Wang, Ziyu and Nakura, Mino and Bhowal, Pratik and Kuo, Chia-Liang and Chen, Yi-Ting and Erickson, Zackory and Held, David},
|
| 29 |
+
booktitle={Robotics: Science and Systems (RSS)},
|
| 30 |
+
year={2025}}
|
|
|
|
| 31 |
```
|