Method Details

Details for method 'QueryInst-Parallel Completion'


Method overview

name QueryInst-Parallel Completion
challenge instance-level semantic labeling
details We propose a novel feature complete network framework queryinst parallel completion. First, the global context module is introduced into the backbone network to obtain instance information. Then, parallel semantic branch and parallel global branch are proposed to extract the semantic information and global information of feature layer, so as to complete the ROI features. In addition, we also propose a feature transfer structure, which explicitly increases the connection between detection and segmentation branches, changes the gradient back-propagation path, and indirectly complements the ROI features.
publication Hai Wang ;Shilin Zhu ;PuPu ;Meng; Le; Apple; Rong
project page / code
used Cityscapes data fine annotations
used external data CoCo
runtime n/a
subsampling no
submission date June, 2022
previous submissions


Average results

Metric Value
AP 35.3755
AP50% 60.8685
AP100m 48.1494
AP50m 50.8495


Class results

Class AP AP50% AP100m AP50m
person 41.4128 72.8727 58.3291 58.4702
rider 31.4676 67.6387 43.6556 43.9216
car 58.4295 83.0294 76.765 79.2953
truck 29.1902 41.1777 38.751 43.4567
bus 43.9733 61.8997 61.6092 68.6619
train 31.6175 53.4349 43.5638 49.8545
motorcycle 24.9731 53.272 31.8189 32.5246
bicycle 21.9403 53.6231 30.7027 30.6113



Download results as .csv file

Benchmark page