Detailled semantic labeling (3D) result


Name Ramiya, A. M. et al. (LCCP_intensity)
Affiliation Indian Institute of Space Science and Technology, India
Abbreviation IIS_5
Strategy (u)nsupervised, (s)upervised, (h)ybrid s

=============================================================
↓ predicted || reference →
 power
low_veg
imp_surf
car
fence_hedge
roof
fac
shrub
tree
missing points
power
0.408
0.000
0.000
0.035
0.000
0.223
---
0.000
0.072
157
low_veg
0.000
0.442
0.343
0.019
0.029
0.002
---
0.143
0.000
2199
imp_surf
0.000
0.023
0.961
0.003
0.002
0.000
---
0.007
0.000
349
car
0.000
0.001
0.056
0.659
0.132
0.000
---
0.072
0.000
297
fence_hedge
0.000
0.004
0.035
0.044
0.295
0.006
---
0.585
0.012
138
roof
0.000
0.005
0.002
0.012
0.016
0.764
---
0.074
0.123
610
fac
0.002
0.021
0.019
0.136
0.079
0.236
---
0.248
0.110
1682
shrub
0.000
0.022
0.049
0.060
0.076
0.027
---
0.714
0.025
701
tree
0.000
0.004
0.006
0.030
0.018
0.066
---
0.345
0.501
1653
Precision/Correctness
 0.814
0.918
0.730
0.225
0.196
0.919
--
0.266
0.638
Recall/Completeness
 0.408
0.442
0.961
0.659
0.295
0.764
--
0.714
0.501
F1
 0.544
0.597
0.830
0.335
0.235
0.834
--
0.387
0.561

Overall accuracy 0.680

Data Download


Download a zip file containing laz files per class. Those can be visualized with CloudCompare

Click here for download!

Data previews (derived from voxelisation at 50cm voxelcube)

Participant labeling (lowest points when multiple labels at one X/Y position)

Red/green image, indicating wrongly classified areas

!!! All red/green visualisations (also the point clouds in the zip file) are defined from the reference point of view.