Method Details


Details for method 'AAF+GLR'

 

Method overview

name AAF+GLR
challenge pixel-level semantic labeling
details
publication Anonymous
project page / code
used Cityscapes data fine annotations
used external data
runtime n/a
subsampling no
submission date July, 2019
previous submissions

 

Average results

Metric Value
IoU Classes 78.2442
iIoU Classes 54.9269
IoU Categories 90.9473
iIoU Categories 79.3083

 

Class results

Class IoU iIoU
road 98.4447 -
sidewalk 84.8036 -
building 92.8368 -
wall 50.845 -
fence 57.7627 -
pole 67.9767 -
traffic light 74.8792 -
traffic sign 78.521 -
vegetation 93.6009 -
terrain 71.5432 -
sky 95.5799 -
person 85.9859 68.7674
rider 68.7995 48.4963
car 95.8395 90.3876
truck 71.3369 35.6408
bus 81.8272 47.1421
train 74.0087 45.8396
motorcycle 66.6573 41.1806
bicycle 75.3902 61.9606

 

Category results

Category IoU iIoU
flat 98.6807 -
nature 93.2621 -
object 73.9127 -
sky 95.5799 -
construction 93.3407 -
human 86.6079 70.3478
vehicle 95.2469 88.2687

 

Links

Download results as .csv file

Benchmark page