Details for method 'RRR-ResNet152-MultiScale'
Method overview
| name |
RRR-ResNet152-MultiScale |
| challenge |
pixel-level semantic labeling |
| details |
update: this submission actually used the coarse labels, which was previously not marked accordingly |
| publication |
Anonymous
|
| project page / code |
|
| used Cityscapes data |
fine annotations, coarse annotations |
| used external data |
ImageNet |
| runtime |
n/a |
| subsampling |
no |
| submission date |
September, 2016 |
| previous submissions |
|
Average results
| Metric |
Value |
| IoU Classes | 75.7876 |
| iIoU Classes | 48.5119 |
| IoU Categories | 89.2502 |
| iIoU Categories | 73.9681 |
Class results
| Class |
IoU |
iIoU |
| road | 98.3166 | - |
| sidewalk | 83.9747 | - |
| building | 92.0444 | - |
| wall | 50.8108 | - |
| fence | 54.501 | - |
| pole | 62.5959 | - |
| traffic light | 67.7161 | - |
| traffic sign | 73.721 | - |
| vegetation | 92.8241 | - |
| terrain | 70.8051 | - |
| sky | 94.9583 | - |
| person | 82.6301 | 60.6025 |
| rider | 60.6455 | 37.9521 |
| car | 94.9582 | 88.4288 |
| truck | 65.2808 | 26.1271 |
| bus | 83.0683 | 39.5569 |
| train | 76.5937 | 43.0835 |
| motorcycle | 63.2536 | 38.7812 |
| bicycle | 71.2663 | 53.5634 |
Category results
| Category |
IoU |
iIoU |
| flat | 98.5059 | - |
| nature | 92.4816 | - |
| object | 68.9638 | - |
| sky | 94.9583 | - |
| construction | 92.3136 | - |
| human | 83.2252 | 61.7922 |
| vehicle | 94.303 | 86.1441 |
Links
Download results as .csv file
Benchmark page
