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, coarse annotations
used external data ImageNet
runtime 1 s
GPU TITAN X
subsampling no
submission date May, 2016
previous submissions 1, 2, 3

 

Average results

Metric Value
IoU Classes 73.187
iIoU Classes 45.3645
IoU Categories 88.0633
iIoU Categories 71.2193

 

Class results

Class IoU iIoU
road 98.414 -
sidewalk 84.7177 -
building 91.5308 -
wall 53.1123 -
fence 52.5725 -
pole 57.5516 -
traffic light 64.8629 -
traffic sign 70.6216 -
vegetation 92.6092 -
terrain 70.4188 -
sky 94.9346 -
person 80.7768 56.803
rider 60.2469 35.0739
car 94.6075 87.0617
truck 60.2778 25.5582
bus 72.5098 39.9619
train 64.5171 36.9347
motorcycle 57.0731 31.2818
bicycle 69.1989 50.2405

 

Category results

Category IoU iIoU
flat 98.5134 -
nature 92.2729 -
object 64.5652 -
sky 94.9346 -
construction 91.7446 -
human 80.8075 57.9774
vehicle 93.6049 84.4612

 

Links

Download results as .csv file

Benchmark page