Frequency-Adaptive Pan-Sharpening with Mixture of Experts
Authors: Xuanhua He, Keyu Yan, Rui Li, Chengjun Xie, Jie Zhang, Man Zhou
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Quantitative and qualitative experiments over multiple datasets demonstrate that our method performs the best against other state-of-the-art ones and comprises a strong generalization ability for real-world scenes. |
| Researcher Affiliation | Academia | 1Hefei Institutes of Physical Science, Chinese Academy of Sciences 2University of Science and Technology of China 3Nanyang Technological University |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks labeled |
| Open Source Code | Yes | Code will be made publicly at https://github.com/alexhe101/FAME-Net. |
| Open Datasets | Yes | Our experiments were conducted on three typical datasets, namely World View-II (WV2), Gaofen2 (GF2), and World View-III (WV3), which include various natural and urban scenarios. Since ground truth is unavailable, we follow the Wald protocol (Wald, Ranchin, and Mangolini 1997) to generate training samples. |
| Dataset Splits | No | The paper mentions generating training samples and evaluating on specific datasets, but does not explicitly detail training, validation, and test dataset splits with percentages or counts. |
| Hardware Specification | Yes | We trained our model using the Python framework on an RTX 3060 GPU, with four experts (n=4), k=2 for each MOE, and a total of 1000 epochs. |
| Software Dependencies | No | The paper mentions using the |
| Experiment Setup | Yes | We trained our model using the Python framework on an RTX 3060 GPU, with four experts (n=4), k=2 for each MOE, and a total of 1000 epochs. We used the Adam optimizer with a learning rate of 5e-4 and linearly decreased the weight of the loss (α) during training. For training samples, we cropped LRMS patches of size 32x32 and PAN patches of size 128x128 from the images, with a batch size of 4. |