Method Details


Details for method 'SAIT'

 

Method overview

name SAIT
challenge pixel-level semantic labeling
details - no post-processing like CRF < onlyHG >
publication Anonymous
project page / code
used Cityscapes data fine annotations, coarse annotations
used external data ImageNet
runtime 1 s
GPU TITAN X
subsampling no
submission date June, 2016
previous submissions 1, 2, 3, 4, 5

 

Average results

Metric Value
IoU Classes 74.0594
iIoU Classes 46.8454
IoU Categories 88.3991
iIoU Categories 72.1252

 

Class results

Class IoU iIoU
road 98.4102 -
sidewalk 84.7932 -
building 91.7147 -
wall 51.5311 -
fence 54.1065 -
pole 59.0607 -
traffic light 65.8914 -
traffic sign 71.8243 -
vegetation 92.6809 -
terrain 71.0932 -
sky 94.8676 -
person 81.2514 58.2697
rider 61.9509 36.1659
car 94.7912 87.266
truck 63.7165 28.6774
bus 76.185 41.7757
train 65.4894 39.7495
motorcycle 57.941 31.9182
bicycle 69.8285 50.9409

 

Category results

Category IoU iIoU
flat 98.5376 -
nature 92.3871 -
object 65.927 -
sky 94.8676 -
construction 91.9654 -
human 81.2792 59.4194
vehicle 93.8301 84.8311

 

Links

Download results as .csv file

Benchmark page