Method Details
Details for method 'kMaX-DeepLab [Cityscapes-fine]'
Method overview
| name | kMaX-DeepLab [Cityscapes-fine] |
| challenge | instance-level semantic labeling |
| details | kMaX-DeepLab w/ ConvNeXt-L backbone (ImageNet-22k + 1k pretrained). This result is obtained by the kMaX-DeepLab trained for Panoptic Segmentation task. No test-time augmentation or other external dataset. |
| publication | k-means Mask Transformer Qihang Yu, Huiyu Wang, Siyuan Qiao, Maxwell Collins, Yukun Zhu, Hartwig Adam, Alan Yuille, and Liang-Chieh Chen ECCV 2022 https://arxiv.org/abs/2207.04044 |
| project page / code | https://github.com/google-research/deeplab2 |
| used Cityscapes data | fine annotations |
| used external data | ImageNet 22k + 1k |
| runtime | n/a |
| subsampling | no |
| submission date | March, 2022 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| AP | 39.7338 |
| AP50% | 61.3181 |
| AP100m | 57.1603 |
| AP50m | 61.2394 |
Class results
| Class | AP | AP50% | AP100m | AP50m |
|---|---|---|---|---|
| person | 36.2101 | 63.5259 | 56.8748 | 57.111 |
| rider | 33.8154 | 64.8719 | 50.559 | 51.3937 |
| car | 55.1987 | 75.2481 | 78.4014 | 81.3537 |
| truck | 40.3789 | 50.9052 | 57.1268 | 65.0529 |
| bus | 53.0865 | 65.8731 | 76.5496 | 85.6468 |
| train | 47.3948 | 64.3953 | 64.8881 | 75.9183 |
| motorcycle | 29.8872 | 56.0491 | 39.7817 | 40.3139 |
| bicycle | 21.8986 | 49.6766 | 33.1013 | 33.1246 |