Method Details


Details for method 'TuSimple_Coarse'

 

Method overview

name TuSimple_Coarse
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 November, 2016
previous submissions 1, 2

 

Average results

Metric Value
IoU Classes 78.3917
iIoU Classes 54.9092
IoU Categories 90.1693
iIoU Categories 75.1406

 

Class results

Class IoU iIoU
road 98.4664 -
sidewalk 85.4617 -
building 92.7922 -
wall 58.3405 -
fence 55.3305 -
pole 65.0955 -
traffic light 73.4632 -
traffic sign 77.9581 -
vegetation 93.303 -
terrain 71.8983 -
sky 95.2037 -
person 84.8053 62.5936
rider 68.5202 43.8306
car 95.3872 88.0934
truck 68.9774 38.1661
bus 84.7421 52.4268
train 79.8096 49.4748
motorcycle 66.0194 43.2911
bicycle 73.8685 61.3973

 

Category results

Category IoU iIoU
flat 98.5813 -
nature 92.9912 -
object 71.9738 -
sky 95.2037 -
construction 93.0023 -
human 84.8011 63.8304
vehicle 94.6316 86.4508

 

Links

Download results as .csv file

Benchmark page