Details for method 'NL_ROI_ROB'
Method overview
| name |
NL_ROI_ROB |
| challenge |
instance-level semantic labeling |
| details |
Non-local ROI on Mask R-CNN |
| publication |
Anonymous
|
| project page / code |
|
| used Cityscapes data |
fine annotations |
| used external data |
ImageNet, COCO, Cityscapes, Kitti, WildDash, ScanNet, ADE20K |
| runtime |
n/a |
| subsampling |
no |
| submission date |
June, 2018 |
| previous submissions |
|
Average results
| Metric |
Value |
| AP | 23.9775 |
| AP50% | 45.7671 |
| AP100m | 36.055 |
| AP50m | 40.811 |
Class results
| Class |
AP | AP50% | AP100m | AP50m |
| person | 28.3309 | 58.1443 | 45.0842 | 45.2449 |
| rider | 20.552 | 54.668 | 31.7639 | 32.5945 |
| car | 45.5458 | 69.9767 | 65.5659 | 69.1025 |
| truck | 22.2841 | 32.1532 | 33.3455 | 41.1969 |
| bus | 30.6404 | 46.2575 | 47.6131 | 59.6901 |
| train | 17.2593 | 33.1294 | 26.1263 | 38.685 |
| motorcycle | 15.0954 | 36.2302 | 19.8773 | 20.7501 |
| bicycle | 12.1122 | 35.5775 | 19.0638 | 19.2241 |
Links
Download results as .csv file
Benchmark page
