Method Details


Details for method 'SAIT'

 

Method overview

name SAIT
challenge pixel-level semantic labeling
details
publication Anonymous
project page / code
used Cityscapes data fine annotations
used external data ImageNet
runtime 1 s
GPU TITAN X
subsampling no
submission date May, 2016
previous submissions 1, 2

 

Average results

Metric Value
IoU Classes 71.5805
iIoU Classes 45.1524
IoU Categories 88.203
iIoU Categories 72.8235

 

Class results

Class IoU iIoU
road 98.2879 -
sidewalk 83.8275 -
building 91.3038 -
wall 42.9349 -
fence 47.9822 -
pole 58.8232 -
traffic light 65.4777 -
traffic sign 71.3841 -
vegetation 92.4334 -
terrain 69.7034 -
sky 94.742 -
person 81.1031 59.9106
rider 59.3572 34.4226
car 94.3022 86.8483
truck 59.5074 25.8809
bus 66.5771 41.4685
train 57.1824 31.5522
motorcycle 55.5114 31.5905
bicycle 69.5881 49.5452

 

Category results

Category IoU iIoU
flat 98.4941 -
nature 92.1959 -
object 65.7248 -
sky 94.742 -
construction 91.6359 -
human 81.199 60.9774
vehicle 93.4295 84.6696

 

Links

Download results as .csv file

Benchmark page