Method Details
Details for method 'DCME'
Method overview
| name | DCME |
| challenge | instance-level semantic labeling |
| details | |
| publication | Distance to Center of Mass Encoding for Instance Segmentation Thomio Watanabe and Denis Wolf 2018 21st International Conference on Intelligent Transportation Systems (ITSC) https://arxiv.org/abs/1711.09060 |
| project page / code | |
| used Cityscapes data | fine annotations |
| used external data | |
| runtime | n/a |
| subsampling | 3 |
| submission date | October, 2017 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| AP | 3.76663 |
| AP50% | 7.73426 |
| AP100m | 6.62281 |
| AP50m | 9.4676 |
Class results
| Class | AP | AP50% | AP100m | AP50m |
|---|---|---|---|---|
| person | 1.7727 | 5.85537 | 3.74859 | 4.06271 |
| rider | 0.710318 | 3.32938 | 1.29459 | 1.3842 |
| car | 15.5283 | 25.6472 | 26.6025 | 35.5387 |
| truck | 1.99661 | 4.02463 | 3.58086 | 5.12316 |
| bus | 4.2955 | 8.30124 | 8.14193 | 14.6595 |
| train | 4.56998 | 9.97649 | 7.73137 | 12.8631 |
| motorcycle | 0.930207 | 3.38729 | 1.3031 | 1.4988 |
| bicycle | 0.329394 | 1.35248 | 0.579558 | 0.610567 |