Method Details
Details for method 'SwiftNetRN-18 ensemble'
Method overview
| name | SwiftNetRN-18 ensemble |
| challenge | pixel-level semantic labeling |
| details | |
| publication | In Defense of Pre-trained ImageNet Architectures for Real-time Semantic Segmentation of Road-driving Images Marin Oršić, Ivan Krešo, Petra Bevandić, Siniša Šegvić CVPR 2019 https://openaccess.thecvf.com/content_CVPR_2019/papers/Orsic_In_Defense_of_Pre-Trained_ImageNet_Architectures_for_Real-Time_Semantic_Segmentation_CVPR_2019_paper.pdf |
| project page / code | https://github.com/orsic/swiftnet |
| used Cityscapes data | fine annotations |
| used external data | ImageNet |
| runtime | n/a |
| subsampling | no |
| submission date | February, 2019 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| IoU Classes | 76.4547 |
| iIoU Classes | 51.4179 |
| IoU Categories | 90.0754 |
| iIoU Categories | 76.495 |
Class results
| Class | IoU | iIoU |
|---|---|---|
| road | 98.5232 | - |
| sidewalk | 85.3919 | - |
| building | 92.507 | - |
| wall | 49.1117 | - |
| fence | 53.8861 | - |
| pole | 64.1871 | - |
| traffic light | 71.1276 | - |
| traffic sign | 76.3169 | - |
| vegetation | 93.0896 | - |
| terrain | 71.0368 | - |
| sky | 95.545 | - |
| person | 84.7798 | 64.96 |
| rider | 67.4689 | 43.5078 |
| car | 95.6487 | 89.0753 |
| truck | 67.5222 | 32.4976 |
| bus | 80.4864 | 44.0287 |
| train | 68.2976 | 41.214 |
| motorcycle | 63.4647 | 37.7642 |
| bicycle | 74.2478 | 58.2952 |
Category results
| Category | IoU | iIoU |
|---|---|---|
| flat | 98.6471 | - |
| nature | 92.777 | - |
| object | 70.6333 | - |
| sky | 95.545 | - |
| construction | 92.771 | - |
| human | 85.1863 | 66.4017 |
| vehicle | 94.9685 | 86.5883 |