Detailled semantic labeling (3D) result


Name Ramiya, A. M. et al. (colorRG_RGB)
Affiliation Indian Institute of Space Science and Technology, India
Abbreviation IIS_3
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
---
0.000
0.000
0.000
0.067
0.867
---
0.040
0.027
0
low_veg
---
0.687
0.104
0.002
0.040
0.019
---
0.137
0.005
501
imp_surf
---
0.159
0.826
0.001
0.003
0.002
---
0.009
0.000
35
car
---
0.051
0.006
0.376
0.205
0.143
---
0.218
0.000
5
fence_hedge
---
0.029
0.012
0.019
0.417
0.048
---
0.464
0.010
9
roof
---
0.002
0.004
0.007
0.009
0.908
---
0.067
0.002
106
fac
---
0.030
0.016
0.017
0.071
0.696
---
0.165
0.001
47
shrub
---
0.066
0.007
0.016
0.176
0.085
---
0.632
0.016
76
tree
---
0.009
0.002
0.001
0.098
0.162
---
0.319
0.407
95
Precision/Correctness
 --
0.778
0.882
0.432
0.157
0.817
--
0.258
0.951
Recall/Completeness
 --
0.687
0.826
0.376
0.417
0.908
--
0.632
0.407
F1
 --
0.730
0.853
0.402
0.228
0.860
--
0.366
0.571

Overall accuracy 0.714

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.