Global Attention Pyramid Network for Semantic Segmentation

Na Zhang,Jun Li,Yongrui Li,Yang Du
DOI: https://doi.org/10.23919/chicc.2019.8865946
2019-01-01
Abstract:Global Attention Pyramid module is a combination of the two pyramid structures and the mechanism of attention. In the paper, we use the two pyramid structures based on the image classification model of deep convolutional neural networks(DCNNs), which taking out the feature maps of different resolutions from the basic image classification model for further processing by attention mechanisms or pyramid pooling structure. The two pyramid structures are used in cooperation with each other. The larger pyramid structure is used to capture the detailed information of the image, while the smaller pyramid structure uses the pyramid pooling module to capture the semantic information of the feature map. Global Attention Pyramid Network(GAPNet) is an end-to-end network without the post-processing of the network such as conditional random fields(CRF). We have obtained excellent results in the Cityscapes datasets. The evaluation metric of results are mean intersection over union(mIoU), and the results of experiment that we got are respectively (67.6/70.7) in test datasets/validation datasets.
What problem does this paper attempt to address?