Details for method 'MaskRCNN_ROB'
Method overview
| name |
MaskRCNN_ROB |
| challenge |
instance-level semantic labeling |
| details |
MaskRCNN Instance segmentation baseline for ROB challenge using default parameters from Matterport's implementation of Mask RCNN
https://github.com/matterport/Mask_RCNN |
| publication |
Anonymous
|
| project page / code |
|
| used Cityscapes data |
fine annotations |
| used external data |
KITTI 2015, Wilddash, ScanNet |
| runtime |
n/a |
| subsampling |
no |
| submission date |
April, 2018 |
| previous submissions |
|
Average results
| Metric |
Value |
| AP | 10.2047 |
| AP50% | 25.2345 |
| AP100m | 14.6144 |
| AP50m | 14.609 |
Class results
| Class |
AP | AP50% | AP100m | AP50m |
| person | 19.1447 | 49.9828 | 29.6454 | 29.251 |
| rider | 10.4755 | 41.7844 | 15.758 | 15.4818 |
| car | 34.1168 | 60.2056 | 48.7462 | 49.6323 |
| truck | 2.74046 | 7.32462 | 2.53644 | 1.60621 |
| bus | 7.17523 | 15.1678 | 11.447 | 12.8116 |
| train | 0 | 0 | 0 | 0 |
| motorcycle | 7.98466 | 27.4104 | 8.78234 | 8.08908 |
| bicycle | 0 | 0 | 0 | 0 |
Links
Download results as .csv file
Benchmark page
