Method Details


Details for method 'SU_Net'

 

Method overview

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

 

Average results

Metric Value
IoU Classes 75.3363
iIoU Classes 52.2935
IoU Categories 88.5454
iIoU Categories 74.974

 

Class results

Class IoU iIoU
road 98.3869 -
sidewalk 84.6544 -
building 91.8717 -
wall 54.2837 -
fence 56.3664 -
pole 54.6034 -
traffic light 67.3585 -
traffic sign 73.7672 -
vegetation 92.6684 -
terrain 71.1205 -
sky 94.5151 -
person 83.7312 62.384
rider 67.3218 42.6548
car 95.189 88.9751
truck 65.0554 35.4831
bus 76.8432 46.5992
train 64.092 43.4384
motorcycle 65.6356 42.3671
bicycle 73.9246 56.446

 

Category results

Category IoU iIoU
flat 98.5389 -
nature 92.3753 -
object 63.6405 -
sky 94.5151 -
construction 92.2782 -
human 84.043 63.3737
vehicle 94.4267 86.5743

 

Links

Download results as .csv file

Benchmark page