MathLee / MCCNet

[TGRS2022] [MCCNet] Multi-Content Complementation Network for Salient Object Detection in Optical Remote Sensing Images

MCCNet

This project provides the code and results for 'Multi-Content Complementation Network for Salient Object Detection in Optical Remote Sensing Images', IEEE TGRS, vol. 60, pp. 1-13, 2022. IEEE link and arxiv link Homepage

Network Architecture

Multi-Content Complementation Module (MCCM)

Requirements

python 2.7 + pytorch 0.4.0 or

python 3.7 + pytorch 1.9.0

Saliency maps

We provide saliency maps and measure results (.mat) (code: i9d0) of all compared methods (code: 5np3) and our MCCNet (code: 3pvq) on ORSSD and EORSSD datasets.

In addition, we also provide saliency maps of our MCCNet (code: 413m) on the recently published ORSI-4199 dataset.

Training

We get the ground truth of edge using sal2edge.m in EGNet，and use data_aug.m for data augmentation.

Modify paths of VGG backbone (code: ego5) and datasets, then run train_MCCNet.py.

Pre-trained model and testing

Download the following pre-trained model, and modify paths of pre-trained model and datasets, then run test_MCCNet.py.

ORSSD (code: awqr)

EORSSD (code: wm3p)

ORSI-4199 (code: 336a)

Evaluation Tool

You can use the evaluation tool (MATLAB version) to evaluate the above saliency maps.

ORSI-SOD_Summary

Citation

    @ARTICLE{Li_2022_MCCNet,
            author = {Gongyang Li and Zhi Liu and Weisi Lin and Haibin Ling},
            title = {Multi-Content Complementation Network for Salient Object Detection in Optical Remote Sensing Images},
            journal = {IEEE Transactions on Geoscience and Remote Sensing},
            volume = {60},
            pages = {1-13},
            year = {2022},
            }

If you encounter any problems with the code, want to report bugs, etc.

Please contact me at [e-mail] or [e-mail].