Method Details


Details for method 'Joint Graph Decomposition and Node Labeling'

 

Method overview

name Joint Graph Decomposition and Node Labeling
challenge instance-level semantic labeling
details
publication Joint Graph Decomposition and Node Labeling: Problem, Algorithms, Applications
Evgeny Levinkov, Jonas Uhrig, Siyu Tang, Mohamed Omran, Eldar Insafutdinov, Alexander Kirillov, Carsten Rother, Thomas Brox, Bernt Schiele, Bjoern Andres
Computer Vision and Pattern Recognition (CVPR) 2017
https://arxiv.org/abs/1611.04399
project page / code
used Cityscapes data fine annotations
used external data ImageNet
runtime n/a
subsampling 8
submission date November, 2016
previous submissions

 

Average results

Metric Value
AP 9.77587
AP50% 23.2022
AP100m 16.8278
AP50m 20.3294

 

Class results

Class AP AP50% AP100m AP50m
person 6.51885 18.352 13.4531 13.9596
rider 9.34098 29.4731 16.5501 17.3574
car 23.0705 38.2558 38.4477 43.9458
truck 6.6722 16.075 11.3348 15.0464
bus 10.8594 21.4584 19.1542 26.0561
train 10.3218 24.5437 16.9326 26.1559
motorcycle 6.80083 21.4217 10.4228 11.5914
bicycle 4.62233 16.0376 8.32726 8.52262

 

Links

Download results as .csv file

Benchmark page