ISPRS BENCHMARK ON
MULTISENSORIAL INDOOR MAPPING AND POSITIONING

Indoor Positioning Dataset

Evaluation

The indoor positioning dataset consists of five data sequences acquired in indoor environments with various complexity. Data sequences of sensor records from smartphones are provided. Users can test their positioning algorithm on these data. The first two sequences (“mimap_in_pose 00” and “mimap_in_pose 01”) were acquired in one building, and the other three sequences (“mimap_in_pose 02”, “mimap_in_pose 03”, “mimap_in_pose 04”) were acquired in another building.
We provide two ways of evaluation as follows:
(1) Evaluation using downloaded ground truth
The ground truth trajectory data of this dataset is centimeter-level accuracy platform trajectory data from the SLAM algorithm. Users can compare their trajectory results to the ground truth trajectory.
(2) Evaluation by submitting results
More ground truth trajectories are available for performance comparison. To participate in the performance comparison, users need to submit the result trajectories generated by their positioning algorithms from the smartphone records. The evaluation results will be listed on the webpage.
Submission data format. In the submitted trajectory file, each line in the trajectory file should be {p_x, p_y, p_z, timestamp (UTC time(s))}. Each ground truth trajectories file (in TXT format) contains an N9 table, where N is the number of frames of this sequence. The format of each row in the file is {frame_id p_x p_y p_z q_x q_y q_z q_w, timestamp}, where frame_id is the index of lidar frame with the current pose, p_x, p_y, and p_z are the translation components of the current pose, q_x, q_y, q_z, and q_w are the quaternion representations of the rotation component of the current pose.
Evaluation Criterion. Our evaluation firstly locates the corresponding pose information in the submitted trajectory results based on the timestamp of each pose in ground truth files. Then, computes translational errors for all possible subsequences of some lengths (5, 10, 25, 50 meters). The evaluation table will rank methods according to the average of translational errors, where errors are measured in percent.

Data Description


Name Size Data description Ground truth
mimap in pose 0016.9 MBA five-floor building scene including data of individual rooms, closed-loop corridors and stairs
mimap in pose 0111.7 MB A three-floor building (same as “mimap in pose 00”)scene including data of individual rooms, closed-loop corridors and stairsPlease submit your results for evaluation
mimap in pose 0210.8 MBA six-floor building scene including data of corridors and stairs
mimap in pose 033.29 MBA single-floor building (same as “mimap in pose 02”) scene including data of multiple roomsPlease submit your results for evaluation
mimap in pose 045.02 MBA single-floor building (same as “mimap in pose 02”) scene including data of multiple roomsPlease submit your results for evaluation

Download


mimap_in_pose_00.zip( 16.9 MB )    [Google]  [Baidu]

mimap_in_pose_01.zip( 11.7 MB )    [Google]  [Baidu]

mimap_in_pose_02.zip( 10.8 MB )    [Google]  [Baidu]

mimap_in_pose_03.zip( 3.29 MB )    [Google]  [Baidu]

mimap_in_pose_04.zip( 5.02 MB )    [Google]  [Baidu]

Download per country

unknown
5871
"United States
754
"Singapore
694
United States
678
nul
503
France
419
"Germany
344
Germany
296
China
118
"Cyprus
80
"France
74
Singapore
72
"Canada
58
"Finland
53
Israel
50
Cyprus
33
"China
19
Netherlands
18
Canada
17
Finland
17
"Netherlands
13
Indonesia
12
"Russia
9
Hong Kong SAR China
8
South Korea
8
Ukraine
8
Ethiopia
5
"Hong Kong SAR China
4
"Indonesia
2
Brazil
2
"South Korea
2
Switzerland
2
Sri Lanka
2
Egypt
2
"Australia
2
United Kingdom
1
"Ireland
1
Thailand
1
India
1
Denmark
1
"Tanzania
1
Spain
1
"Dominican Republic
1
Japan
1
"United Kingdom
1
Australia
1
Russia
1

Copyright

The MiMAP benchmark is published under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License (https://creativecommons.org/licenses/by-nc-sa/3.0/).You must attribute the work in the manner specified by the authors, you may not use this work for commercial purposes and if you alter, transform, or build upon this work, you may distribute the resulting work only under the same license. Contact us if you are interested in commercial usage.

Citation

If you use MiMAP benchmark, please cite both the following papers:

  • C. Wen, Y. Dai, Y. Xia, Y. Lian, C. Wang, J. Li, Towards Efficient 3-D Colored Mapping in GPS/GNSS-denied Environments, IEEE Geoscience and Remote Sensing Letters, 17, 147-151, 2020.

  • C. Wang, S. Hou, C. Wen, Z. Gong, Q. Li, X. Sun, J. Li, Semantic Line Framework-based Indoor Building Modeling using Backpacked Laser Scanning Point Cloud, ISPRS Journal of Photogrammetry and Remote Sensing, Vol. 143, pp. 150-166, 2018.