Details for method 'GMIS: Graph Merge for Instance Segmentation'
Method overview
| name |
GMIS: Graph Merge for Instance Segmentation |
| challenge |
instance-level semantic labeling |
| details |
|
| publication |
Yiding Liu, Siyu Yang, Bin Li, Wengang Zhou, Jizheng Xu, Houqiang Li, Yan Lu
|
| project page / code |
|
| used Cityscapes data |
fine annotations, coarse annotations |
| used external data |
ImageNet |
| runtime |
n/a |
| subsampling |
no |
| submission date |
February, 2018 |
| previous submissions |
|
Average results
| Metric |
Value |
| AP | 27.6337 |
| AP50% | 44.6342 |
| AP100m | 42.7077 |
| AP50m | 47.8861 |
Class results
| Class |
AP | AP50% | AP100m | AP50m |
| person | 29.277 | 50.6091 | 47.8604 | 47.8532 |
| rider | 24.0727 | 47.3748 | 37.9872 | 38.4224 |
| car | 42.7178 | 56.941 | 66.3331 | 70.5974 |
| truck | 25.3737 | 33.9259 | 39.4846 | 48.0353 |
| bus | 37.2477 | 47.0152 | 60.7813 | 75.0953 |
| train | 32.9191 | 52.7874 | 47.1876 | 59.7225 |
| motorcycle | 17.5915 | 39.2336 | 23.9431 | 25.0953 |
| bicycle | 11.8699 | 29.1867 | 18.0847 | 18.2674 |
Links
Download results as .csv file
Benchmark page
