TorWIC-SLAM
Notes on the Dataset
- The provided results of image-based semantic segmentation are not very precise.
Detailed Information
| General Information | |
| Name | TorWIC-SLAM |
| Release Year | 2023 |
| Terms of Use | Unclear (Citation) |
| Access Requirements | None |
| Dataset Size | 440.7 GB | Partial download is possible, i.e. the data is split into several files (e.g., epochs or data types) |
| Documentation | In-depth documentation of acquisition and characteristics of the dataset, e.g., via an explicit dataset paper or a comprehensive multi-page metadata document |
| Code | Yes |
| Applications | Long-term localization and mapping |
| Detailed Applications | Long-term mapping |
| Acquisition | |
| Number of Scenes | 3 |
| Number of Epochs per Scene (minimum) | 5 |
| Number of Epochs per Scene (median) | 5 |
| Number of Epochs per Scene (maximum) | 10 |
| General Scene Type | Indoor |
| Specific Scene Type | Warehouse route |
| Location | Toronto (Canada) |
| Acquisition Type | Mobile laser scanning | Depth camera mounted on vehicle |
| Acquisition Device | Ouster OS1-128 LiDAR + Microsoft Azure Kinect RGB-D camera |
| Acquisition Platform | OTTO 100 robot |
| Scan Interval | Minutes to Months |
| Acquisition Months | Janurary: 0 February: 0 March: 0 April: 0 May: 0 June: 14 July: 0 August: 0 September: 0 October: 6 November: 0 December: 0 |
| Representation | |
| Data Representation | Unstructured, local (e.g., local point clouds or laser scans with poses) |
| Specific Data Representation | Local point clouds |
| File Format/Encoding | PCD |
| Raw Data | ROS bags |
| Additional Data | RGBD | GT pointcloud |
| Coordinate System | m |
| Quality and Usability | |
| Registration | Finely registered |
| Number of Partial Epochs | 23.53% |
| Unusable Data Reason | - |
| Splits | No |
| Per-Point Attributes | |
| Intensity/Reflectivity | No |
| Color | Yes | Color is not naturally included, but images are available that could be backprojected |
| Semantic Labels | 16 | The labels can be computed from other data/mapped to the point cloud, i.e., it can not trivially be done during point cloud construction |
| Instance Labels | No |
| Change Labels | - |
| Statistics | |
| Number of Points per Epoch (minimum) | 240M |
| Number of Points per Epoch (median) | 400M |
| Number of Points per Epoch (maximum) | 666M |
| Avg. Point Spacing (minimum) | 2.1mm |
| Avg. Point Spacing (median) | 4.1mm |
| Avg. Point Spacing (maximum) | 4.6mm |
| Avg. Change Points per Epoch | - |
Paper Reference 1
@inproceedings{qian2023datasettorwicslam,
author = {Qian, Jingxing and Chatrath, Veronica and Servos, James and Mavrinac, Aaron and Burgard, Wolfram and Waslander, Steven L. and Schoellig, Angela},
booktitle = {Proc. Robotics: Science and Systems XIX},
title = {POV-SLAM: Probabilistic Object-Aware Variational SLAM in Semi-Static Environments},
year = {2023},
doi = {10.15607/RSS.2023.XIX.069},
}