Method Details


Details for method 'VlocNet++_ROB'

 

Method overview

name VlocNet++_ROB
challenge pixel-level semantic labeling
details
publication Anonymous
project page / code
used Cityscapes data
used external data
runtime n/a
subsampling no
submission date May, 2018
previous submissions

 

Average results

Metric Value
IoU Classes 62.6834
iIoU Classes 33.8919
IoU Categories 83.3613
iIoU Categories 60.7557

 

Class results

Class IoU iIoU
road 97.4009 -
sidewalk 78.0725 -
building 88.3816 -
wall 37.7761 -
fence 28.8365 -
pole 45.8818 -
traffic light 48.8038 -
traffic sign 57.0747 -
vegetation 89.2899 -
terrain 66.1587 -
sky 92.9465 -
person 72.2261 40.8071
rider 50.2976 23.9428
car 91.6843 82.4483
truck 47.0789 18.1316
bus 56.0062 29.9654
train 41.4871 16.0993
motorcycle 42.581 21.1216
bicycle 59.0005 38.6194

 

Category results

Category IoU iIoU
flat 98.0559 -
nature 89.3032 -
object 51.6105 -
sky 92.9465 -
construction 88.3747 -
human 72.8803 42.2406
vehicle 90.3578 79.2708

 

Links

Download results as .csv file

Benchmark page