Method Details


Details for method 'MIX6D_RVC'

 

Method overview

name MIX6D_RVC
challenge pixel-level semantic labeling
details MIX6D_RVC
publication Anonymous
project page / code
used Cityscapes data fine annotations
used external data ImageNet, SCANNET, VIPER, MVD, ADE20K, Wilddash
runtime n/a
subsampling no
submission date September, 2022
previous submissions 1

 

Average results

Metric Value
IoU Classes 78.5814
iIoU Classes 56.9484
IoU Categories 88.7672
iIoU Categories 74.9043

 

Class results

Class IoU iIoU
road 97.9195 -
sidewalk 82.152 -
building 92.172 -
wall 62.6256 -
fence 61.9134 -
pole 59.5808 -
traffic light 70.0011 -
traffic sign 73.0098 -
vegetation 93.0049 -
terrain 69.7542 -
sky 94.7185 -
person 82.9648 65.0972
rider 65.9049 44.1196
car 94.6638 85.3803
truck 77.1436 43.649
bus 90.2936 55.9107
train 84.971 57.2966
motorcycle 67.2949 46.1783
bicycle 72.9578 57.9556

 

Category results

Category IoU iIoU
flat 97.671 -
nature 92.4792 -
object 66.5909 -
sky 94.7185 -
construction 92.1739 -
human 83.5685 66.2131
vehicle 94.1683 83.5954

 

Links

Download results as .csv file

Benchmark page