Packaging
Description
For this dataset, it was recorded the task of stacking the completed boxed televisions on wooden pallets and wrapping them in a plastic membrane for shipping. Three healthy men participated in the MoCap recording session, two stacking medium-size TVs and one stacking large-size TVs. Their average age was 31.5±6.2 years, their height was 167.8±4.6 cm, and their average weight of 65.3±9.9 kg.
Name convention and segmentation
Preparing TVs for Shipping (TVP & TVL):
- P01: Placing TVs on a wooden pallet (bottom level)
- P02: Preparing to wrap the bottom level with a membrane
- P03: Wrapping the bottom level
- P04: Placing TVs on top of the bottom level (second level)
- P05: Placing TVs on top of the second level (third level)
- P06: Wrapping the second level with a plastic membrane
- P07: Wrapping the third level with a plastic membrane
- P08: Placing TVs on top of the third level (fourth level)
- P09: Wrapping the fourth level with a plastic membrane
Medium-size TVs: P08
Medium-size TVs: P07
Large-size TVs: P01
Metrics
Gesture recognition
| Method | Motion descriptors | Number of classes | Accuracy | F1-score |
|---|---|---|---|---|
| HMMs | Joint angles from all 52 IMUs | 5 | 0.950 | 0.916 |
| HMMs | Joint angles from: RFA and H | 5 | 0.888 | 0.850 |
| HMMs | Joint angles from: RSH1, LFA, SP2 | 5 | 0.901 | 0.843 |
* Hidden Markov Models (HMMs).
Motion generation
| Method | MAE | RMSE | Average U1 |
|---|---|---|---|
| KF-GOM | 7.947 | 15.978 | 0.195 |
| KF-RGOM | 9.867 | 20.937 | 0.125 |
| VAE-RGOM | 0.213 | 3.231 | 0.019 |
| ATT-RGOM | 0.398 | 3.339 | 0.025 |