Method Details

Details for method 'EaNet-V1'


Method overview

name EaNet-V1
challenge pixel-level semantic labeling
details Parsing very high resolution (VHR) urban scene images into regions with semantic meaning, e.g. buildings and cars, is a fundamental task necessary for interpreting and understanding urban scenes. However, due to the huge quantity of details contained in an image and the large variations of objects in scale and appearance, the existing semantic segmentation methods often break one object into pieces, or confuse adjacent objects and thus fail to depict these objects consistently. To address this issue, we propose a concise and effective edge-aware neural network (EaNet) for urban scene semantic segmentation. The proposed EaNet model is deployed as a standard balanced encoder-decoder framework. Specifically, we devised two plug-and-play modules that append on top of the encoder and decoder respectively, i.e., the large kernel pyramid pooling (LKPP) and the edge-aware loss (EA loss) function, to extend the model ability in learning discriminating features. The LKPP module captures rich multi-scale context with strong continuous feature relations to promote coherent labeling of multi-scale urban objects. The EA loss module learns edge information directly from semantic segmentation prediction, which avoids costly post-processing or extra edge detection. During training, EA loss imposes a strong geometric awareness to guide object structure learning at both the pixel- and image-level, and thus effectively separates confusing objects with sharp contours.
publication Parsing Very High Resolution Urban Scene Images by Learning Deep ConvNets with Edge-Aware Loss
Xianwei Zheng, Linxi Huan, Gui-Song Xia, Jianya Gong
project page / code
used Cityscapes data fine annotations
used external data ImageNet
runtime n/a
subsampling no
submission date May, 2020
previous submissions


Average results

Metric Value
IoU Classes 81.6836
iIoU Classes 59.6017
IoU Categories 91.1883
iIoU Categories 77.8342


Class results

Class IoU iIoU
road 98.7604 -
sidewalk 87.1909 -
building 93.7706 -
wall 67.2685 -
fence 64.1345 -
pole 67.467 -
traffic light 75.6267 -
traffic sign 79.8804 -
vegetation 93.7732 -
terrain 72.4115 -
sky 95.6646 -
person 86.9444 65.8753
rider 73.4417 49.4382
car 96.045 90.746
truck 80.3559 44.8858
bus 89.425 54.8061
train 80.5463 58.9446
motorcycle 71.7323 48.827
bicycle 77.5501 63.2907


Category results

Category IoU iIoU
flat 98.7694 -
nature 93.5023 -
object 73.8175 -
sky 95.6646 -
construction 93.8809 -
human 86.846 66.7768
vehicle 95.8373 88.8916



Download results as .csv file

Benchmark page