Method Details


Details for method 'firenet'

 

Method overview

name firenet
challenge pixel-level semantic labeling
details
publication Anonymous
project page / code
used Cityscapes data fine annotations
used external data
runtime n/a
subsampling 2
submission date September, 2017
previous submissions

 

Average results

Metric Value
IoU Classes 68.2208
iIoU Classes 47.7792
IoU Categories 84.867
iIoU Categories 75.4618

 

Class results

Class IoU iIoU
road 94.103 -
sidewalk 74.2461 -
building 87.4149 -
wall 40.1201 -
fence 44.6396 -
pole 54.1547 -
traffic light 65.4224 -
traffic sign 65.1033 -
vegetation 90.0187 -
terrain 66.4925 -
sky 92.1428 -
person 76.7398 64.858
rider 61.8175 40.1127
car 92.7801 85.9118
truck 45.0219 27.6178
bus 64.8548 40.5497
train 59.2712 31.6821
motorcycle 54.3801 37.4279
bicycle 67.4716 54.0739

 

Category results

Category IoU iIoU
flat 95.0299 -
nature 89.7507 -
object 60.6021 -
sky 92.1428 -
construction 87.2328 -
human 77.6251 66.4182
vehicle 91.6857 84.5053

 

Links

Download results as .csv file

Benchmark page