Method Details
Details for method 'ResNet-38'
Method overview
| name | ResNet-38 |
| challenge | pixel-level semantic labeling |
| details | single model, single scale, no post-processing with CRFs |
| publication | Wider or Deeper: Revisiting the ResNet Model for Visual Recognition Anonymous https://arxiv.org/abs/1611.10080 |
| project page / code | https://github.com/itijyou/ademxapp |
| used Cityscapes data | fine annotations, coarse annotations |
| used external data | ImageNet, Places |
| runtime | n/a |
| subsampling | no |
| submission date | November, 2016 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| IoU Classes | 79.9278 |
| iIoU Classes | 59.7081 |
| IoU Categories | 91.2055 |
| iIoU Categories | 80.7875 |
Class results
| Class | IoU | iIoU |
|---|---|---|
| road | 98.6335 | - |
| sidewalk | 86.6736 | - |
| building | 93.3192 | - |
| wall | 59.3031 | - |
| fence | 63.2344 | - |
| pole | 68.3067 | - |
| traffic light | 75.8876 | - |
| traffic sign | 79.6732 | - |
| vegetation | 93.8549 | - |
| terrain | 73.7356 | - |
| sky | 95.4602 | - |
| person | 86.9335 | 71.1304 |
| rider | 70.4968 | 51.4612 |
| car | 96.0039 | 90.8102 |
| truck | 69.8636 | 43.1435 |
| bus | 83.4743 | 54.6121 |
| train | 76.5065 | 51.5828 |
| motorcycle | 70.0831 | 49.5377 |
| bicycle | 77.1853 | 65.3868 |
Category results
| Category | IoU | iIoU |
|---|---|---|
| flat | 98.737 | - |
| nature | 93.5086 | - |
| object | 74.4309 | - |
| sky | 95.4602 | - |
| construction | 93.7078 | - |
| human | 87.2108 | 72.462 |
| vehicle | 95.3835 | 89.1131 |