Object Change Detection
Notes on the Dataset
- During writing the paper, the download link for the dataset stopped working (timeout). It is unclear whether the dataset was removed or is only temporarily unavailable. In any case, another dataset that uses the same raw scans is available here. The point clouds differ very slightly and are split into surfaces. Also the first epoch is not available. However, ROS bags for all scans are provided.
Detailed Information
| General Information | |
| Name | Object Change Detection |
| Release Year | 2020 |
| Terms of Use | Unclear |
| Access Requirements | None |
| Dataset Size | 142.2 MB | Partial download is possible, i.e. the data is split into several files (e.g., epochs or data types) |
| Documentation | Multi-paragraph description of the dataset, e.g., in the form of a paper section or a comprehensive readme file |
| Code | No |
| Applications | Dynamic object detection, localization, and modelling |
| Detailed Applications | Novel object detection |
| Acquisition | |
| Number of Scenes | 5 |
| Number of Epochs per Scene (minimum) | 6 |
| Number of Epochs per Scene (median) | 6 |
| Number of Epochs per Scene (maximum) | 7 |
| General Scene Type | Indoor |
| Specific Scene Type | Room |
| Location | Unclear |
| Acquisition Type | Depth camera mounted on vehicle |
| Acquisition Device | Asus Xtion PRO Live |
| Acquisition Platform | Human support robot |
| Scan Interval | Minutes | Educated guess by the authors |
| Acquisition Months | |
| Representation | |
| Data Representation | Unstructured, globally aligned (e.g., point cloud or ray cloud) |
| Specific Data Representation | Point cloud |
| File Format/Encoding | PCD |
| Raw Data | - |
| Additional Data | List of points belonging to novel objects |
| Coordinate System | m |
| Quality and Usability | |
| Registration | Between finely and coarsely registered. The reviewers disagreed on this |
| Number of Partial Epochs | - |
| Unusable Data Reason | - |
| Splits | No |
| Per-Point Attributes | |
| Intensity/Reflectivity | No |
| Color | Yes |
| Semantic Labels | - |
| Instance Labels | No |
| Change Labels | 2 |
| Statistics | |
| Number of Points per Epoch (minimum) | 52K |
| Number of Points per Epoch (median) | 206K |
| Number of Points per Epoch (maximum) | 761K |
| Avg. Point Spacing (minimum) | 5.0mm |
| Avg. Point Spacing (median) | 7.2mm |
| Avg. Point Spacing (maximum) | 7.7mm |
| Avg. Change Points per Epoch | 2.66% |
Paper Reference 1
@inproceedings{langer2020datasetocd,
author = {Langer, Edith and Patten, Timothy and Vincze, Markus},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
title = {Robust and Efficient Object Change Detection by Combining Global Semantic Information and Local Geometric Verification},
year = {2020},
pages = {8453--8460},
doi = {10.1109/IROS45743.2020.9341664},
}