Details for method 'MaskRCNN_BOSH'
Method overview
| name |
MaskRCNN_BOSH |
| challenge |
instance-level semantic labeling |
| details |
MaskRCNN segmentation baseline for Bosh autodrive challenge ,
using Matterport's implementation of Mask RCNN https://github.com/matterport/Mask_RCNN
55k iterations, default parameters (backbone :resenet 101)
19hours for training |
| publication |
Jin shengtao, Yi zhihao, Liu wei [Our team name is firefly]
|
| project page / code |
|
| used Cityscapes data |
fine annotations |
| used external data |
coco |
| runtime |
n/a |
| subsampling |
no |
| submission date |
June, 2018 |
| previous submissions |
|
Average results
| Metric |
Value |
| AP | 12.7945 |
| AP50% | 28.0393 |
| AP100m | 22.0956 |
| AP50m | 26.738 |
Class results
| Class |
AP | AP50% | AP100m | AP50m |
| person | 14.2652 | 34.16 | 28.5647 | 29.492 |
| rider | 9.37836 | 31.803 | 16.6462 | 17.4244 |
| car | 25.1719 | 46.857 | 40.2545 | 43.6854 |
| truck | 10.9419 | 16.8288 | 18.4578 | 25.2685 |
| bus | 15.1569 | 25.0958 | 26.8238 | 41.1039 |
| train | 13.6321 | 26.0632 | 22.9875 | 32.4816 |
| motorcycle | 7.55356 | 22.5612 | 11.5499 | 12.6775 |
| bicycle | 6.25593 | 20.9453 | 11.4803 | 11.7706 |
Links
Download results as .csv file
Benchmark page
