Method Details
Details for method 'PolygonRNN++'
Method overview
| name | PolygonRNN++ |
| challenge | instance-level semantic labeling |
| details | |
| publication | Efficient Annotation of Segmentation Datasets with Polygon-RNN++ D. Acuna, H. Ling, A. Kar, and S. Fidler CVPR 2018 https://arxiv.org/abs/1803.09693 |
| project page / code | http://www.cs.toronto.edu/polyrnn/ |
| used Cityscapes data | fine annotations |
| used external data | ImageNet |
| runtime | n/a |
| subsampling | no |
| submission date | February, 2018 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| AP | 25.4856 |
| AP50% | 45.4693 |
| AP100m | 39.3141 |
| AP50m | 43.4099 |
Class results
| Class | AP | AP50% | AP100m | AP50m |
|---|---|---|---|---|
| person | 29.3598 | 54.9815 | 49.6394 | 50.086 |
| rider | 21.7923 | 49.2541 | 34.5972 | 35.1537 |
| car | 48.2971 | 69.9941 | 69.3161 | 72.6777 |
| truck | 21.1291 | 29.7407 | 32.3922 | 41.2953 |
| bus | 32.3369 | 47.4636 | 52.4865 | 62.1108 |
| train | 23.7278 | 41.4335 | 36.0809 | 44.7549 |
| motorcycle | 13.6035 | 34.0887 | 18.2702 | 19.1054 |
| bicycle | 13.6385 | 36.7985 | 21.7301 | 22.0958 |