Plaza
- 1 minPlaza
Data for this category is collected in two large plazas, featuring highly-dynamic pedestrian, changing appearance, and ground reflections, which present substantial challenges for SLAM algorithms. Specifically, we record data across multiple floors and lift to evaluate relocation performance.
| Sequence | Duration (s) | Distance (m) | Difficulty | Description | Bag | Ground Truth |
|---|---|---|---|---|---|---|
| plaza_01 | 195.5 | 253.5 | ⭐ | 1 floor, medium dynamic people, glass and ground reflection, returns to origin. | OneDrive 百度网盘 | OneDrive 百度网盘 |
| plaza_02 | 535.8 | 761.1 | ⭐⭐ | 1 floor, high dynamic people, transparent glass, ground reflections, larger scale, returns to origin. | OneDrive 百度网盘 | OneDrive 百度网盘 |
| plaza_03 | 1381.9 | 1838.4 | ⭐⭐⭐⭐ | 7 floor, accessed via elevators, multi-level indoor scenes, returns to origin using a lift, limited visibility, environmental changes, medium dynamic people, returns to origin. | OneDrive 百度网盘 | OneDrive 百度网盘 |
| plaza_04 | 2215.9 | 3713.8 | ⭐⭐⭐⭐ | 7 floors, accessed via elevators, multi-level indoor scenes, returns to origin using a lift, limited visibility, environmental changes, high dynamic people, transparent glass, ground reflections, returns to origin. | OneDrive 百度网盘 | OneDrive 百度网盘 |