Method Details
Details for method 'DeepLab LargeFOV Strong'
Method overview
| name | DeepLab LargeFOV Strong |
| challenge | pixel-level semantic labeling |
| details | Trained on a pre-release version of the dataset |
| publication | Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs L.-C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, and A. L. Yuille ICLR 2015 http://arxiv.org/pdf/1412.7062v3.pdf |
| project page / code | https://bitbucket.org/deeplab/deeplab-public/ |
| used Cityscapes data | fine annotations |
| used external data | ImageNet |
| runtime | 4 s |
| subsampling | 2 |
| submission date | April, 2016 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| IoU Classes | 63.0737 |
| iIoU Classes | 34.4913 |
| IoU Categories | 81.2262 |
| iIoU Categories | 58.6945 |
Class results
| Class | IoU | iIoU |
|---|---|---|
| road | 97.2603 | - |
| sidewalk | 77.659 | - |
| building | 87.721 | - |
| wall | 43.6283 | - |
| fence | 40.4511 | - |
| pole | 29.7182 | - |
| traffic light | 44.5199 | - |
| traffic sign | 55.384 | - |
| vegetation | 89.3613 | - |
| terrain | 67.007 | - |
| sky | 92.6764 | - |
| person | 71.1619 | 40.4936 |
| rider | 49.3977 | 23.3488 |
| car | 91.3723 | 78.8148 |
| truck | 48.7132 | 20.3022 |
| bus | 56.7157 | 31.8562 |
| train | 49.1485 | 24.7749 |
| motorcycle | 47.9382 | 21.1202 |
| bicycle | 58.5656 | 35.2195 |
Category results
| Category | IoU | iIoU |
|---|---|---|
| flat | 97.7513 | - |
| nature | 89.0307 | - |
| object | 40.4435 | - |
| sky | 92.6764 | - |
| construction | 88.0148 | - |
| human | 71.0116 | 41.2574 |
| vehicle | 89.6549 | 76.1315 |