Method Details
Details for method 'ERINet'
Method overview
| name | ERINet |
| challenge | pixel-level semantic labeling |
| details | Efficient residual inception networks for real-time semantic segmentation |
| publication | Anonymous |
| project page / code | |
| used Cityscapes data | fine annotations |
| used external data | |
| runtime | 0.023 s Titan X |
| subsampling | 2 |
| submission date | August, 2018 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| IoU Classes | 69.8371 |
| iIoU Classes | 44.0896 |
| IoU Categories | 87.3586 |
| iIoU Categories | 73.4406 |
Class results
| Class | IoU | iIoU |
|---|---|---|
| road | 97.9834 | - |
| sidewalk | 82.3909 | - |
| building | 90.4311 | - |
| wall | 41.6955 | - |
| fence | 47.6951 | - |
| pole | 58.9612 | - |
| traffic light | 63.0308 | - |
| traffic sign | 67.6908 | - |
| vegetation | 91.9603 | - |
| terrain | 69.5719 | - |
| sky | 94.673 | - |
| person | 78.2959 | 59.7014 |
| rider | 57.3259 | 35.4369 |
| car | 93.428 | 87.6049 |
| truck | 50.4608 | 23.1798 |
| bus | 60.7392 | 31.9348 |
| train | 61.5339 | 32.0538 |
| motorcycle | 53.3165 | 33.2315 |
| bicycle | 65.72 | 49.5733 |
Category results
| Category | IoU | iIoU |
|---|---|---|
| flat | 98.2405 | - |
| nature | 91.6173 | - |
| object | 64.864 | - |
| sky | 94.673 | - |
| construction | 90.5683 | - |
| human | 79.2295 | 61.5133 |
| vehicle | 92.3177 | 85.3679 |