Details for method 'AInnoSegmentation'
Method overview
| name |
AInnoSegmentation |
| challenge |
instance-level semantic labeling |
| details |
AInnoSegmentation use SE-Resnet 152 as backbone and FPN model to extract multi-level features and use self-develop method to combine multi-features and use COCO datasets to pre-train model and so on |
| publication |
Faen Zhang, Jiahong Wu, Haotian Cao, Zhizheng Yang, Jianfei Song, Ze Huang, Jiashui Huang, Shenglan Ben
|
| project page / code |
|
| used Cityscapes data |
fine annotations |
| used external data |
MSCOCO |
| runtime |
n/a |
| subsampling |
no |
| submission date |
September, 2019 |
| previous submissions |
|
Average results
| Metric |
Value |
| AP | 39.5284 |
| AP50% | 65.9604 |
| AP100m | 53.9121 |
| AP50m | 56.729 |
Class results
| Class |
AP | AP50% | AP100m | AP50m |
| person | 42.3331 | 75.637 | 59.6439 | 59.891 |
| rider | 32.5505 | 69.446 | 46.7345 | 47.4681 |
| car | 57.5507 | 83.2891 | 76.4319 | 78.9246 |
| truck | 39.9609 | 53.658 | 53.6292 | 58.4637 |
| bus | 51.3375 | 69.9959 | 70.81 | 80.6934 |
| train | 39.8002 | 62.0322 | 52.8239 | 56.5484 |
| motorcycle | 30.5567 | 59.7555 | 38.7892 | 39.2663 |
| bicycle | 22.1375 | 53.8693 | 32.4338 | 32.5766 |
Links
Download results as .csv file
Benchmark page
