Method Details
Details for method 'Instance Annotation'
Method overview
| name | Instance Annotation |
| challenge | instance-level semantic labeling |
| details | Based on DCME |
| publication | Instance Segmentation as Image Segmentation Annotation Thomio Watanabe and Denis F. Wolf 2019 IEEE Intelligent Vehicles Symposium (IV) https://arxiv.org/abs/1902.05498 |
| project page / code | |
| used Cityscapes data | fine annotations |
| used external data | ImageNet |
| runtime | 4.416 s GPU |
| subsampling | 2 |
| submission date | January, 2019 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| AP | 7.72195 |
| AP50% | 14.9459 |
| AP100m | 13.63 |
| AP50m | 16.5973 |
Class results
| Class | AP | AP50% | AP100m | AP50m |
|---|---|---|---|---|
| person | 6.65838 | 17.055 | 14.1057 | 15.297 |
| rider | 3.08833 | 8.81521 | 5.57873 | 5.96581 |
| car | 24.1363 | 38.097 | 41.3619 | 50.5918 |
| truck | 6.01641 | 10.6556 | 10.1248 | 11.6033 |
| bus | 9.76437 | 15.1314 | 17.6895 | 22.522 |
| train | 6.40944 | 12.685 | 10.8868 | 16.5211 |
| motorcycle | 3.62415 | 10.6648 | 5.54371 | 6.35592 |
| bicycle | 2.07827 | 6.46309 | 3.74883 | 3.9212 |