Method Details


Details for method 'mogo_semantic'

 

Method overview

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

 

Average results

Metric Value
IoU Classes 85.3333
iIoU Classes 69.6158
IoU Categories 93.1541
iIoU Categories 85.1927

 

Class results

Class IoU iIoU
road 98.9633 -
sidewalk 89.0113 -
building 94.8657 -
wall 71.278 -
fence 68.1153 -
pole 76.1238 -
traffic light 82.3188 -
traffic sign 85.0096 -
vegetation 94.4439 -
terrain 74.5217 -
sky 96.2956 -
person 90.1705 78.1386
rider 79.4137 63.3376
car 96.9091 92.5399
truck 82.1565 56.4327
bus 95.5283 69.105
train 88.615 64.5111
motorcycle 76.1273 60.5619
bicycle 81.4645 72.2998

 

Category results

Category IoU iIoU
flat 98.8855 -
nature 94.2075 -
object 81.0041 -
sky 96.2956 -
construction 94.896 -
human 90.1787 79.1067
vehicle 96.6116 91.2787

 

Links

Download results as .csv file

Benchmark page