Method Details
Details for method 'Panoptic-DeepLab [Mapillary Vistas]'
Method overview
| name | Panoptic-DeepLab [Mapillary Vistas] |
| challenge | pixel-level semantic labeling |
| details | We employ a stronger backbone, WR-41, in Panoptic-DeepLab. For Panoptic-DeepLab, please refer to https://arxiv.org/abs/1911.10194. For wide-ResNet-41 (WR-41) backbone, please refer to https://arxiv.org/abs/2005.10266. |
| publication | Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation Bowen Cheng, Maxwell D. Collins, Yukun Zhu, Ting Liu, Thomas S. Huang, Hartwig Adam, Liang-Chieh Chen https://arxiv.org/abs/1911.10194 |
| project page / code | |
| used Cityscapes data | fine annotations |
| used external data | ImageNet, Mapillary Vistas Research Edition. |
| runtime | n/a |
| subsampling | no |
| submission date | May, 2020 |
| previous submissions | 1 |
Average results
| Metric | Value |
|---|---|
| IoU Classes | 84.5419 |
| iIoU Classes | 68.6516 |
| IoU Categories | 92.8505 |
| iIoU Categories | 82.8412 |
Class results
| Class | IoU | iIoU |
|---|---|---|
| road | 98.8465 | - |
| sidewalk | 88.4043 | - |
| building | 94.4479 | - |
| wall | 64.3122 | - |
| fence | 68.3437 | - |
| pole | 75.3238 | - |
| traffic light | 81.0451 | - |
| traffic sign | 84.1963 | - |
| vegetation | 94.1942 | - |
| terrain | 73.7355 | - |
| sky | 96.0953 | - |
| person | 89.6959 | 75.8285 |
| rider | 78.6468 | 61.9026 |
| car | 96.7378 | 90.0563 |
| truck | 82.1915 | 57.65 |
| bus | 93.6982 | 69.4838 |
| train | 90.1646 | 64.1424 |
| motorcycle | 76.4094 | 61.7365 |
| bicycle | 79.8066 | 68.4126 |
Category results
| Category | IoU | iIoU |
|---|---|---|
| flat | 98.8177 | - |
| nature | 93.9254 | - |
| object | 80.2398 | - |
| sky | 96.0953 | - |
| construction | 94.7286 | - |
| human | 89.7238 | 76.7678 |
| vehicle | 96.4226 | 88.9146 |