Method Details
Details for method 'Panoptic-DeepLab w/ SWideRNet [Cityscapes-fine]'
Method overview
| name | Panoptic-DeepLab w/ SWideRNet [Cityscapes-fine] |
| challenge | panoptic semantic labeling |
| details | We revisit the architecture design of Wide Residual Networks. We design a baseline model by incorporating the simple and effective Squeeze-and-Excitation and Switchable Atrous Convolution to the Wide-ResNets. Its network capacity is further scaled up or down by adjusting the width (i.e., channel size) and depth (i.e., number of layers), resulting in a family of SWideRNets (short for Scaling Wide Residual Networks). We demonstrate that such a simple scaling scheme, coupled with grid search, identifies several SWideRNets that significantly advance state-of-the-art performance on panoptic segmentation datasets in both the fast model regime and strong model regime. |
| publication | Scaling Wide Residual Networks for Panoptic Segmentation Liang-Chieh Chen, Huiyu Wang, Siyuan Qiao https://arxiv.org/abs/2011.11675 |
| project page / code | |
| used Cityscapes data | fine annotations |
| used external data | ImageNet |
| runtime | n/a |
| subsampling | no |
| submission date | October, 2020 |
| previous submissions |
Average results
| Metric | All | Things | Stuff |
|---|---|---|---|
| PQ | 64.8392 | 56.4913 | 70.9104 |
| SQ | 83.3507 | 81.7117 | 84.5427 |
| RQ | 77.0361 | 69.2236 | 82.7179 |
Class results
| Class | PQ | SQ | RQ |
|---|---|---|---|
| road | 98.5517 | 98.6818 | 99.8682 |
| sidewalk | 79.2669 | 85.985 | 92.1869 |
| building | 89.4529 | 91.6018 | 97.6542 |
| wall | 39.5326 | 76.7747 | 51.4917 |
| fence | 47.0453 | 76.8948 | 61.1814 |
| pole | 66.5272 | 72.8958 | 91.2634 |
| traffic light | 58.6024 | 79.1967 | 73.996 |
| traffic sign | 73.251 | 82.468 | 88.8235 |
| vegetation | 91.2302 | 92.1707 | 98.9796 |
| terrain | 45.8893 | 79.7352 | 57.5521 |
| sky | 90.6647 | 93.5649 | 96.9004 |
| person | 58.4195 | 78.4877 | 74.4315 |
| rider | 56.3141 | 75.4021 | 74.6851 |
| car | 70.6637 | 85.449 | 82.697 |
| truck | 51.2731 | 87.8289 | 58.3784 |
| bus | 61.961 | 89.1787 | 69.4796 |
| train | 53.4564 | 86.3527 | 61.9048 |
| motorcycle | 51.6199 | 77.1768 | 66.8852 |
| bicycle | 48.2228 | 73.8175 | 65.327 |