Method Details


Details for method 'NAVINFO_DLR'

 

Method overview

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

 

Average results

Metric Value
IoU Classes 82.7817
iIoU Classes 63.1054
IoU Categories 91.8781
iIoU Categories 82.176

 

Class results

Class IoU iIoU
road 98.6864 -
sidewalk 87.1504 -
building 93.7918 -
wall 59.9676 -
fence 63.2617 -
pole 70.948 -
traffic light 77.5665 -
traffic sign 81.7738 -
vegetation 93.8674 -
terrain 72.8995 -
sky 95.5516 -
person 88.1546 73.5623
rider 74.899 54.8529
car 96.3483 91.1808
truck 81.6799 48.0816
bus 93.5607 58.143
train 90.4117 58.8996
motorcycle 73.7451 54.0549
bicycle 78.5886 66.068

 

Category results

Category IoU iIoU
flat 98.7442 -
nature 93.5565 -
object 76.7554 -
sky 95.5516 -
construction 94.0158 -
human 88.4487 74.6919
vehicle 96.0745 89.6602

 

Links

Download results as .csv file

Benchmark page