OpenLORIS-Scene
Notes on the Dataset
- The depth images are quite noisy and the ground truth poses are not perfect (as also mentioned in a Github issue and the paper itself).
- Some scenes have only little overlap between the epochs (e.g., the office scene).
Detailed Information
| General Information | |
| Name | OpenLORIS-Scene |
| Release Year | 2020 |
| Terms of Use | CC BY-ND 4.0 |
| Access Requirements | None |
| Dataset Size | 207.1 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 | Lifelong SLAM |
| Acquisition | |
| Number of Scenes | 5 |
| Number of Epochs per Scene (minimum) | 2 |
| Number of Epochs per Scene (median) | 5 |
| Number of Epochs per Scene (maximum) | 7 |
| General Scene Type | Indoor |
| Specific Scene Type | Rooms | market |
| Location | Unclear |
| Acquisition Type | Depth camera mounted on vehicle |
| Acquisition Device | RealSense D435i RGB-D camera and RoboSense RS-LiDAR-16 |
| Acquisition Platform | Segway Deliverybot S1 robot and Gaussian Scrubber 75 robot |
| Scan Interval | Minutes to Hours |
| Acquisition Months | Janurary: 0 February: 0 March: 0 April: 0 May: 0 June: 29 July: 3 August: 0 September: 0 October: 0 November: 0 December: 0 |
| Representation | |
| Data Representation | Structured, local (e.g., RGBD or range images with poses (and intrinsics) |
| Specific Data Representation | Color and depth images |
| File Format/Encoding | PNG |
| Raw Data | ROS bags |
| Additional Data | Stereo fisheye images | 2D LiDAR |
| Coordinate System | m |
| Quality and Usability | |
| Registration | Coarsely registered |
| Number of Partial Epochs | 70.59% |
| Unusable Data Reason | - |
| Splits | No |
| Per-Point Attributes | |
| Intensity/Reflectivity | No |
| Color | Yes |
| Semantic Labels | - |
| Instance Labels | No |
| Change Labels | - |
| Statistics | |
| Number of Points per Epoch (minimum) | 96M |
| Number of Points per Epoch (median) | 536M |
| Number of Points per Epoch (maximum) | 2B |
| Avg. Point Spacing (minimum) | 761µm |
| Avg. Point Spacing (median) | 1.3mm |
| Avg. Point Spacing (maximum) | 2.9mm |
| Avg. Change Points per Epoch | - |
Paper Reference 1
@inproceedings{shi2020datasetopenloris,
author = {Xuesong Shi and
Dongjiang Li and
Pengpeng Zhao and
Qinbin Tian and
Yuxin Tian and
Qiwei Long and
Chunhao Zhu and
Jingwei Song and
Fei Qiao and
Le Song and
Yangquan Guo and
Zhigang Wang and
Yimin Zhang and
Baoxing Qin and
Wei Yang and
Fangshi Wang and
Rosa H. M. Chan and
Qi She},
title = {Are We Ready for Service Robots? The OpenLORIS-Scene Datasets for
Lifelong SLAM},
booktitle = {IEEE International Conference on Robotics and Automation, ICRA},
pages = {3139--3145},
year = {2020},
doi = {10.1109/ICRA40945.2020.9196638},
}