Method Details
Details for method 'Pixelwise Instance Segmentation with a Dynamically Instantiated Network'
Method overview
| name | Pixelwise Instance Segmentation with a Dynamically Instantiated Network |
| challenge | panoptic semantic labeling |
| details | Results are produced using the method from our CVPR 2017 paper, "Pixelwise Instance Segmentation with a Dynamically Instantiated Network." On the instance segmentation benchmark, the identical model achieved a mean AP of 23.4 This model also served as the fully supervised baseline in our ECCV 2018 paper, "Weakly- and Semi-Supervised Panoptic Segmentation". |
| publication | Pixelwise Instance Segmentation with a Dynamically Instantiated Network Anurag Arnab and Philip H.S Torr Computer Vision and Pattern Recognition (CVPR) 2017 http://www.robots.ox.ac.uk/~aarnab/instances_dynamic_network.html |
| project page / code | |
| used Cityscapes data | fine annotations, coarse annotations |
| used external data | ImageNet |
| runtime | n/a |
| subsampling | no |
| submission date | July, 2019 |
| previous submissions |
Average results
| Metric | All | Things | Stuff |
|---|---|---|---|
| PQ | 55.4058 | 44.0417 | 63.6707 |
| SQ | 79.7325 | 77.3133 | 81.4919 |
| RQ | 68.0579 | 57.0002 | 76.0998 |
Class results
| Class | PQ | SQ | RQ |
|---|---|---|---|
| road | 98.2951 | 98.4249 | 99.8682 |
| sidewalk | 75.1595 | 84.339 | 89.1159 |
| building | 87.5609 | 90.2125 | 97.0608 |
| wall | 31.2084 | 73.9782 | 42.186 |
| fence | 35.7204 | 73.7353 | 48.4441 |
| pole | 43.3234 | 66.3181 | 65.3266 |
| traffic light | 47.6683 | 72.2752 | 65.9539 |
| traffic sign | 65.0791 | 75.7843 | 85.8741 |
| vegetation | 89.5595 | 90.9164 | 98.5075 |
| terrain | 38.0613 | 77.8806 | 48.8713 |
| sky | 88.7417 | 92.5459 | 95.8894 |
| person | 44.6547 | 74.4191 | 60.0043 |
| rider | 43.1112 | 70.8425 | 60.855 |
| car | 50.7514 | 78.4839 | 64.6647 |
| truck | 40.5601 | 84.2889 | 48.1203 |
| bus | 49.9158 | 85.1573 | 58.616 |
| train | 46.24 | 81.3482 | 56.8421 |
| motorcycle | 42.4285 | 73.8725 | 57.4347 |
| bicycle | 34.6719 | 70.0942 | 49.4647 |