/A2-FPN

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

A2-FPN for Semantic Segmentation of Fine-Resolution Remotely Sensed Images

⭐ Welcome to my HomePage ⭐

In this repository, we implement the Feature Pyramid Network (FPN) with Attention Aggregation Module (AAM), i.e., A2-FPN, for semantic segmentation of fine-resolution remotely sensed images.

The detailed results can be seen in the A2-FPN for Semantic Segmentation of Fine-Resolution Remotely Sensed Images.

The related repositories include:

  • ABCNet->An efficient segmentation model.
  • MACU-Net->A modified version of U-Net.
  • BANet->A Transformer-based segmentation network.
  • MAResU-Net->A ResNet-based network with attention mechanism.
  • Multi-Attention-Network->A network with multi kernel attention mechanism.

If our code is helpful to you, please cite:

Rui Li, Libo Wang, Ce Zhang, Chenxi Duan & Shunyi Zheng (2022) A2-FPN for semantic segmentation of fine-resolution remotely sensed images, International Journal of Remote Sensing, 43:3, 1131-1155, DOI: 10.1080/01431161.2022.2030071.

Network:

network
Fig. 1. The overall architecture of A2-FPN.

Result:

The result on the UAVid dataset can seen from here or download by this link:

Method Backbone building tree clutter road vegetation static car moving car human mIoU
MSD - 79.8 74.5 57.0 74.0 55.9 32.1 62.9 19.7 57.0
BiSeNet ResNet-18 85.7 78.3 64.7 61.1 77.3 63.4 48.6 17.5 61.5
SwiftNet ResNet-18 85.3 78.2 64.1 61.5 76.4 62.1 51.1 15.7 61.1
ShelfNet ResNet-18 85.3 78.2 44.1 61.4 43.4 21.0 52.6 3.6 47.0
MANet ResNet-18 85.4 77.0 64.5 77.8 60.3 53.6 67.2 14.9 62.6
BANet ResNet-18 85.4 78.9 66.6 80.7 62.1 52.8 69.3 21.0 64.6
ABCNet ResNet-18 86.4 79.9 67.4 81.2 63.1 48.4 69.8 13.9 63.8
A2-FPN ResNet-18 87.2 80.1 67.4 80.2 63.7 53.3 70.1 23.4 65.7

Result
Fig. 2. The experimental results on the UAVid test set. The first column illustrates the input RGB images, the second column depicts the outputs of MSD and the third column shows the predictions of our A2-FPN.