Method Details
Details for method 'SCMNet coarse'
Method overview
| name | SCMNet coarse |
| challenge | pixel-level semantic labeling |
| details | |
| publication | SCMNet: Shared Context Mining Network for Real-time Semantic Segmentation Tanmay Singha; Moritz Bergemann; Duc-Son Pham; Aneesh Krishna 2021 Digital Image Computing: Techniques and Applications (DICTA) https://ieeexplore.ieee.org/abstract/document/9647401 |
| project page / code | https://github.com/tanmaysingha/SCMNet |
| used Cityscapes data | fine annotations, coarse annotations |
| used external data | |
| runtime | n/a |
| subsampling | no |
| submission date | November, 2021 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| IoU Classes | 68.2979 |
| iIoU Classes | 38.2758 |
| IoU Categories | 87.1734 |
| iIoU Categories | 69.1289 |
Class results
| Class | IoU | iIoU |
|---|---|---|
| road | 97.9264 | - |
| sidewalk | 81.6245 | - |
| building | 90.481 | - |
| wall | 39.7923 | - |
| fence | 44.7254 | - |
| pole | 57.0626 | - |
| traffic light | 62.199 | - |
| traffic sign | 68.7317 | - |
| vegetation | 91.9318 | - |
| terrain | 67.393 | - |
| sky | 94.6752 | - |
| person | 77.9892 | 53.6861 |
| rider | 56.1441 | 28.6623 |
| car | 92.9328 | 85.602 |
| truck | 52.402 | 18.1574 |
| bus | 62.0123 | 31.1652 |
| train | 51.7832 | 23.1716 |
| motorcycle | 44.1341 | 20.4354 |
| bicycle | 63.7199 | 45.3266 |
Category results
| Category | IoU | iIoU |
|---|---|---|
| flat | 98.2759 | - |
| nature | 91.6725 | - |
| object | 63.4626 | - |
| sky | 94.6752 | - |
| construction | 90.8679 | - |
| human | 79.0798 | 55.0781 |
| vehicle | 92.1796 | 83.1798 |