Method Details
Details for method 'PMSDSEN'
Method overview
| name | PMSDSEN |
| challenge | pixel-level semantic labeling |
| details | MM '23: Proceedings of the 31th ACM International Conference on Multimedia |
| publication | Efficient Parallel Multi-Scale Detail and Semantic Encoding Network for Lightweight Semantic Segmentation Xiao Liu, Xiuya Shi, Lufei Chen, Linbo Qing, Chao Ren ACM International Conference on Multimedia 2023 https://dl.acm.org/doi/abs/10.1145/3581783.3611848 |
| project page / code | https://github.com/liux520/PMSDSEN |
| used Cityscapes data | fine annotations |
| used external data | |
| runtime | n/a |
| subsampling | no |
| submission date | November, 2023 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| IoU Classes | 73.9923 |
| iIoU Classes | 49.8708 |
| IoU Categories | 88.7791 |
| iIoU Categories | 75.6981 |
Class results
| Class | IoU | iIoU |
|---|---|---|
| road | 98.1634 | - |
| sidewalk | 83.9045 | - |
| building | 91.4072 | - |
| wall | 44.3689 | - |
| fence | 53.1443 | - |
| pole | 62.8432 | - |
| traffic light | 67.4151 | - |
| traffic sign | 72.6269 | - |
| vegetation | 92.6158 | - |
| terrain | 70.1919 | - |
| sky | 94.8203 | - |
| person | 81.2284 | 63.7012 |
| rider | 63.0866 | 40.6056 |
| car | 94.3988 | 88.0539 |
| truck | 65.2159 | 29.4389 |
| bus | 75.2499 | 45.0403 |
| train | 66.9815 | 43.8035 |
| motorcycle | 58.069 | 34.63 |
| bicycle | 70.1215 | 53.6932 |
Category results
| Category | IoU | iIoU |
|---|---|---|
| flat | 98.3859 | - |
| nature | 92.3457 | - |
| object | 68.7326 | - |
| sky | 94.8203 | - |
| construction | 91.8287 | - |
| human | 81.6448 | 65.2536 |
| vehicle | 93.6954 | 86.1425 |