Method Details


Details for method 'GEELY-ATC-SEG'

 

Method overview

name GEELY-ATC-SEG
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 October, 2023
previous submissions

 

Average results

Metric Value
IoU Classes 86.6564
iIoU Classes 73.9023
IoU Categories 93.2516
iIoU Categories 85.3967

 

Class results

Class IoU iIoU
road 98.9029 -
sidewalk 89.2476 -
building 95.0417 -
wall 74.0156 -
fence 73.771 -
pole 76.3152 -
traffic light 81.4647 -
traffic sign 84.9201 -
vegetation 94.6279 -
terrain 75.5528 -
sky 96.3186 -
person 90.2805 79.0612
rider 80.3618 65.1018
car 96.8439 91.2769
truck 86.9312 66.8006
bus 95.7158 74.8442
train 93.3194 73.4744
motorcycle 80.3855 66.8286
bicycle 82.4552 73.8307

 

Category results

Category IoU iIoU
flat 98.8934 -
nature 94.4183 -
object 81.1363 -
sky 96.3186 -
construction 95.1513 -
human 90.3201 80.0301
vehicle 96.5231 90.7633

 

Links

Download results as .csv file

Benchmark page