Method Details


Details for method 'RRL'

 

Method overview

name RRL
challenge instance-level semantic labeling
details
publication Anonymous
project page / code
used Cityscapes data fine annotations, stereo
used external data ImageNet
runtime n/a
subsampling no
submission date October, 2017
previous submissions

 

Average results

Metric Value
AP 29.6673
AP50% 56.0806
AP100m 40.8692
AP50m 41.8176

 

Class results

Class AP AP50% AP100m AP50m
person 33.8463 67.1427 49.661 49.4097
rider 26.8958 63.7363 38.3746 38.3516
car 51.9073 78.3657 69.3202 70.7856
truck 24.1832 35.7866 31.3399 32.0225
bus 35.5776 54.3341 49.6067 53.6258
train 25.2954 49.5546 34.5448 36.6726
motorcycle 20.9262 50.881 26.5361 26.4875
bicycle 18.7066 48.8437 27.5701 27.1852

 

Links

Download results as .csv file

Benchmark page