Method Details
Details for method 'NVSegNet'
Method overview
| name | NVSegNet |
| challenge | pixel-level semantic labeling |
| details | train on downsampled images (2x on each side), training takes 20 hours. The model in evaluation only trained on GTfine train |
| publication | Anonymous |
| project page / code | |
| used Cityscapes data | fine annotations |
| used external data | ImageNet |
| runtime | 0.43 s 1 Nvidia Titan X, Intel i7 |
| subsampling | 2 |
| submission date | April, 2016 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| IoU Classes | 63.4053 |
| iIoU Classes | 35.5009 |
| IoU Categories | 84.7057 |
| iIoU Categories | 64.2389 |
Class results
| Class | IoU | iIoU |
|---|---|---|
| road | 97.4402 | - |
| sidewalk | 77.5417 | - |
| building | 88.6301 | - |
| wall | 38.3513 | - |
| fence | 33.5123 | - |
| pole | 50.6061 | - |
| traffic light | 52.5005 | - |
| traffic sign | 60.7586 | - |
| vegetation | 90.4138 | - |
| terrain | 66.8702 | - |
| sky | 93.7976 | - |
| person | 74.2953 | 46.1374 |
| rider | 51.2657 | 24.5849 |
| car | 91.8306 | 83.219 |
| truck | 37.5805 | 19.3021 |
| bus | 46.5186 | 27.2532 |
| train | 44.3251 | 22.7302 |
| motorcycle | 46.818 | 20.0501 |
| bicycle | 61.6453 | 40.7304 |
Category results
| Category | IoU | iIoU |
|---|---|---|
| flat | 98.0555 | - |
| nature | 90.0296 | - |
| object | 57.1359 | - |
| sky | 93.7976 | - |
| construction | 88.5185 | - |
| human | 74.9589 | 47.2582 |
| vehicle | 90.4436 | 81.2196 |