Details for method 'NV-ADLR'
Method overview
| name |
NV-ADLR |
| challenge |
pixel-level semantic labeling |
| details |
|
| publication |
NVIDIA Applied Deep Learning Research
|
| project page / code |
|
| used Cityscapes data |
fine annotations, coarse annotations |
| used external data |
ImageNet |
| runtime |
n/a |
| subsampling |
no |
| submission date |
April, 2018 |
| previous submissions |
|
Average results
| Metric |
Value |
| IoU Classes | 81.9997 |
| iIoU Classes | 62.4678 |
| IoU Categories | 91.8537 |
| iIoU Categories | 81.5313 |
Class results
| Class |
IoU |
iIoU |
| road | 98.6849 | - |
| sidewalk | 87.1259 | - |
| building | 93.8032 | - |
| wall | 61.1916 | - |
| fence | 63.8039 | - |
| pole | 70.6129 | - |
| traffic light | 77.6124 | - |
| traffic sign | 81.3275 | - |
| vegetation | 94.1139 | - |
| terrain | 74.4014 | - |
| sky | 95.9911 | - |
| person | 87.6319 | 71.9842 |
| rider | 72.6147 | 52.3726 |
| car | 96.2075 | 92.0011 |
| truck | 73.0739 | 45.0324 |
| bus | 92.4352 | 60.1706 |
| train | 88.4978 | 60.7929 |
| motorcycle | 70.6398 | 51.3511 |
| bicycle | 78.2242 | 66.0379 |
Category results
| Category |
IoU |
iIoU |
| flat | 98.7966 | - |
| nature | 93.7904 | - |
| object | 76.4762 | - |
| sky | 95.9911 | - |
| construction | 94.1681 | - |
| human | 87.7822 | 72.7195 |
| vehicle | 95.9713 | 90.343 |
Links
Download results as .csv file
Benchmark page
