Method Details


Details for method 'SSNet'

 

Method overview

name SSNet
challenge pixel-level semantic labeling
details
publication Anonymous
project page / code
used Cityscapes data fine annotations, stereo
used external data
runtime n/a
subsampling no
submission date June, 2022
previous submissions

 

Average results

Metric Value
IoU Classes 72.4569
iIoU Classes 51.6832
IoU Categories 89.8929
iIoU Categories 78.2051

 

Class results

Class IoU iIoU
road 97.9209 -
sidewalk 81.7155 -
building 92.1423 -
wall 44.9307 -
fence 50.6749 -
pole 65.0469 -
traffic light 72.2528 -
traffic sign 76.5934 -
vegetation 92.5791 -
terrain 68.0059 -
sky 95.074 -
person 84.8385 67.56
rider 65.3312 48.0064
car 94.6773 88.6417
truck 52.2488 32.1131
bus 63.7498 43.908
train 47.0459 31.6791
motorcycle 58.2837 42.5622
bicycle 73.5692 58.9947

 

Category results

Category IoU iIoU
flat 98.3971 -
nature 92.267 -
object 71.6913 -
sky 95.074 -
construction 92.2081 -
human 85.3385 69.2455
vehicle 94.2742 87.1647

 

Links

Download results as .csv file

Benchmark page