Method Details


Details for method 'GGCF'

 

Method overview

name GGCF
challenge pixel-level semantic labeling
details
publication Anonymous
project page / code
used Cityscapes data fine annotations, coarse annotations
used external data ImageNet,Mapillary
runtime n/a
subsampling no
submission date March, 2019
previous submissions

 

Average results

Metric Value
IoU Classes 83.2098
iIoU Classes 63.0042
IoU Categories 91.9828
iIoU Categories 81.2667

 

Class results

Class IoU iIoU
road 98.8095 -
sidewalk 87.8396 -
building 94.1025 -
wall 66.0048 -
fence 66.1357 -
pole 71.1062 -
traffic light 78.4315 -
traffic sign 82.2445 -
vegetation 94.0931 -
terrain 74.5068 -
sky 95.7947 -
person 88.0753 72.1935
rider 74.0314 54.2096
car 96.4678 91.1476
truck 79.9362 50.1512
bus 92.4186 59.4206
train 90.8238 59.6997
motorcycle 71.776 51.5641
bicycle 78.3881 65.6475

 

Category results

Category IoU iIoU
flat 98.7894 -
nature 93.7689 -
object 76.9901 -
sky 95.7947 -
construction 94.3158 -
human 88.2279 73.0947
vehicle 95.9931 89.4388

 

Links

Download results as .csv file

Benchmark page