Method Details
Details for method 'LiteSeg-Shufflenet'
Method overview
| name | LiteSeg-Shufflenet |
| challenge | pixel-level semantic labeling |
| details | |
| publication | LiteSeg: A Litewiegth ConvNet for Semantic Segmentation Taha Emara, Hossam E. Abd El Munim, Hazem M. Abbas DICTA 2019 https://arxiv.org/abs/1912.06683 |
| project page / code | https://github.com/tahaemara/LiteSeg |
| used Cityscapes data | fine annotations, coarse annotations |
| used external data | |
| runtime | 0.007518 s Intel Core i7-8700 @ 3.2GHZ, 16GB memory, and NVIDIA GTX1080Ti GPU card. Input image resolution 360X640. |
| subsampling | no |
| submission date | January, 2019 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| IoU Classes | 65.1725 |
| iIoU Classes | 41.0075 |
| IoU Categories | 85.3995 |
| iIoU Categories | 67.2767 |
Class results
| Class | IoU | iIoU |
|---|---|---|
| road | 97.0818 | - |
| sidewalk | 77.8785 | - |
| building | 89.4798 | - |
| wall | 41.8202 | - |
| fence | 42.5491 | - |
| pole | 49.8056 | - |
| traffic light | 52.9251 | - |
| traffic sign | 65.5972 | - |
| vegetation | 91.4841 | - |
| terrain | 67.7983 | - |
| sky | 93.9913 | - |
| person | 76.1197 | 52.6705 |
| rider | 50.1027 | 33.2528 |
| car | 91.4479 | 81.7381 |
| truck | 43.3791 | 22.748 |
| bus | 51.7873 | 33.6613 |
| train | 48.0271 | 30.6558 |
| motorcycle | 44.337 | 26.5861 |
| bicycle | 62.6648 | 46.7478 |
Category results
| Category | IoU | iIoU |
|---|---|---|
| flat | 97.8841 | - |
| nature | 91.2236 | - |
| object | 57.4361 | - |
| sky | 93.9913 | - |
| construction | 89.6993 | - |
| human | 77.277 | 55.0089 |
| vehicle | 90.2848 | 79.5445 |