Method Details


Details for method 'Valeo DAR Germany'

 

Method overview

name Valeo DAR Germany
challenge pixel-level semantic labeling
details Valeo DAR Germany, New Algo Lab
publication Anonymous
project page / code
used Cityscapes data fine annotations, coarse annotations
used external data ImageNet, COCO
runtime n/a
subsampling no
submission date June, 2019
previous submissions

 

Average results

Metric Value
IoU Classes 82.8313
iIoU Classes 62.9463
IoU Categories 91.9947
iIoU Categories 82.1537

 

Class results

Class IoU iIoU
road 98.666 -
sidewalk 86.9374 -
building 93.8327 -
wall 58.5811 -
fence 63.5422 -
pole 70.9809 -
traffic light 78.1596 -
traffic sign 82.2475 -
vegetation 94.0301 -
terrain 73.2137 -
sky 95.4429 -
person 88.5011 73.3099
rider 74.4827 55.072
car 96.4712 91.6135
truck 81.2402 47.2755
bus 93.6741 58.5564
train 90.5234 57.8097
motorcycle 74.2125 53.1786
bicycle 79.0551 66.755

 

Category results

Category IoU iIoU
flat 98.7619 -
nature 93.6642 -
object 76.8527 -
sky 95.4429 -
construction 94.1301 -
human 88.8969 74.3654
vehicle 96.2145 89.9421

 

Links

Download results as .csv file

Benchmark page