Details for method 'SegNet'
Method overview
| name |
SegNet |
| challenge |
instance-level semantic labeling |
| details |
|
| publication |
Anonymous
|
| project page / code |
|
| used Cityscapes data |
fine annotations, coarse annotations |
| used external data |
ImageNet |
| runtime |
0.5 s
|
| subsampling |
no |
| submission date |
September, 2017 |
| previous submissions |
|
Average results
| Metric |
Value |
| AP | 29.5401 |
| AP50% | 55.6056 |
| AP100m | 43.1996 |
| AP50m | 45.7778 |
Class results
| Class |
AP | AP50% | AP100m | AP50m |
| person | 29.9487 | 57.9132 | 48.1097 | 47.9946 |
| rider | 23.4252 | 56.4513 | 35.0719 | 35.2494 |
| car | 43.4303 | 67.462 | 61.4302 | 62.5237 |
| truck | 29.8157 | 46.2116 | 43.1517 | 50.7892 |
| bus | 41.0492 | 65.0887 | 61.7293 | 69.9542 |
| train | 33.2796 | 63.0423 | 44.5642 | 48.1136 |
| motorcycle | 18.7142 | 43.8577 | 26.1388 | 26.3003 |
| bicycle | 16.6577 | 44.8179 | 25.4011 | 25.2974 |
Links
Download results as .csv file
Benchmark page
