Details for method 'EnsembleModel_Bosch'
Method overview
| name |
EnsembleModel_Bosch |
| challenge |
pixel-level semantic labeling |
| details |
we've ensembled three model(erfnet,deeplab-mobilenet,tusimple) and gained 0.57 improvment of IoU Classes value. The best single model is 73.8549 |
| publication |
Jin shengtao, Yi zhihao, Liu wei [Our team name was MaskRCNN_BOSH,firefly]
|
| project page / code |
|
| used Cityscapes data |
fine annotations |
| used external data |
|
| runtime |
n/a |
| subsampling |
no |
| submission date |
June, 2018 |
| previous submissions |
|
Average results
| Metric |
Value |
| IoU Classes | 74.4298 |
| iIoU Classes | 48.9372 |
| IoU Categories | 88.4858 |
| iIoU Categories | 72.9178 |
Class results
| Class |
IoU |
iIoU |
| road | 98.163 | - |
| sidewalk | 83.5814 | - |
| building | 91.4158 | - |
| wall | 48.453 | - |
| fence | 53.6917 | - |
| pole | 59.311 | - |
| traffic light | 66.759 | - |
| traffic sign | 71.8555 | - |
| vegetation | 92.5554 | - |
| terrain | 70.8547 | - |
| sky | 94.5156 | - |
| person | 82.3673 | 60.3573 |
| rider | 64.4913 | 41.0748 |
| car | 94.7307 | 86.2299 |
| truck | 61.7588 | 30.6445 |
| bus | 77.4998 | 44.0171 |
| train | 71.1988 | 40.0528 |
| motorcycle | 59.8874 | 33.7174 |
| bicycle | 71.0756 | 55.404 |
Category results
| Category |
IoU |
iIoU |
| flat | 98.4864 | - |
| nature | 92.191 | - |
| object | 66.0381 | - |
| sky | 94.5156 | - |
| construction | 91.8405 | - |
| human | 82.6146 | 61.7447 |
| vehicle | 93.7144 | 84.0909 |
Links
Download results as .csv file
Benchmark page
