The python code implementation of the AAAI 2024 paper "LDS$^2$AE: Local Diffusion Shared-Specific Autoencoder for Multimodal Remote Sensing Image Classification with Arbitrary Missing Modalities"
We have presented test cases of the proposed model in config.py file.
The windowsize is set to 11 for Trento and Berlin，and is set to 27 for Houston2013！
The train_num_perclass is set to 40.
The optimizer is Adam.
The more detailed training settings are shown in experiments of this paper.
Training & Testing just run the LDS2AE_main.py
@inproceedings{Qu2024LDS2AELD,
title={LDS2AE: Local Diffusion Shared-Specific Autoencoder for Multimodal Remote Sensing Image Classification with Arbitrary Missing Modalities},
author={Jiahui Qu and Yuanbo Yang and Wenqian Dong and Yufei Yang},
booktitle={AAAI Conference on Artificial Intelligence},
year={2024},
url={https://api.semanticscholar.org/CorpusID:268701740} }