Method Details

Details for method 'CASIA_IVA_DANet_NoCoarse'


Method overview

name CASIA_IVA_DANet_NoCoarse
challenge pixel-level semantic labeling
details we address the scene segmentation task by capturing rich contextual dependencies based on the selfattention mechanism. Unlike previous works that capture contexts by multi-scale features fusion, we propose a Dual Attention Networks (DANet) to adaptively integrate local features with their global dependencies. Specifically, we append two types of attention modules on top of traditional dilated FCN, which model the semantic interdependencies in spatial and channel dimensions respectively. The position attention module selectively aggregates the features at each position by a weighted sum of the features at all positions. Similar features would be related to each other regardless of their distances. Meanwhile, the channel attention module selectively emphasizes interdependent channel maps by integrating associated features among all channel maps. We sum the outputs of the two attention modules to further improve feature representation which contributes to more precise segmentation results
publication Dual Attention Network for Scene Segmentation
Jun Fu, Jing Liu, Haijie Tian, Yong Li, Yongjun Bao, Zhiwei Fang,and Hanqing Lu
project page / code
used Cityscapes data fine annotations
used external data ImageNet
runtime n/a
subsampling no
submission date September, 2018
previous submissions


Average results

Metric Value
IoU Classes 81.4715
iIoU Classes 62.2662
IoU Categories 91.6012
iIoU Categories 82.6392


Class results

Class IoU iIoU
road 98.6072 -
sidewalk 86.1454 -
building 93.482 -
wall 56.1653 -
fence 63.2564 -
pole 69.6762 -
traffic light 77.2663 -
traffic sign 81.2635 -
vegetation 93.8529 -
terrain 72.898 -
sky 95.7008 -
person 87.2603 73.8281
rider 72.9164 54.6445
car 96.2488 91.9385
truck 76.8086 46.2346
bus 89.4933 60.3187
train 86.5048 52.4974
motorcycle 72.2191 51.9428
bicycle 78.1938 66.7248


Category results

Category IoU iIoU
flat 98.7303 -
nature 93.5116 -
object 75.8037 -
sky 95.7008 -
construction 93.9431 -
human 87.6742 74.774
vehicle 95.845 90.5045



Download results as .csv file

Benchmark page