Upload new introductory README
Browse files
README.md
CHANGED
|
@@ -1,3 +1,10 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# WearMoCap Dataset
|
| 2 |
+
|
| 3 |
+
This is the training and test data for WearMoCap, a system facilitating multimodal pose tracking from smartwatch and smartphone data. The library source code and tutorials are available on [__github__](https://github.com/wearable-motion-capture). This published technical report details data collection and dataset composition
|
| 4 |
+
[__WearMoCap: multimodal pose tracking for ubiquitous robot control using a smartwatch__](https://www.frontiersin.org/journals/robotics-and-ai/articles/10.3389/frobt.2024.1478016/full). A video is available [__here__](https://www.youtube.com/watch?v=hyQY5dyWPLU).
|
| 5 |
+
|
| 6 |
+
## Dataset Format
|
| 7 |
+
|
| 8 |
+
We collected data for three modes, each represented as one subfolder in this dataset: __Watch Only__, __Phone Uarm__, and __Phone Pocket__. The subfolders contain a README with additional information.
|
| 9 |
+
|
| 10 |
+
In general, we collected ground truth optical motion capture data for left arm motions and body orientation and synced it with smartwatch and phone sensor measurements. This dataset contains the synced, calibrated, and preprocessed data as CSVs.
|