Dual-Window Multiscale Transformer for Hyperspectral Snapshot Compressive Imaging
Authors: Fulin Luo, Xi Chen, Xiuwen Gong, Weiwen Wu, Tan Guo
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on both the simulated and the real data are conducted to demonstrate the superior performance, stability, and generalization ability of our DWMT. |
| Researcher Affiliation | Academia | 1College of Computer Science, Chongqing University 2Faculty of Engineering, The University of Sydney 3Department of Biomedical Engineering, Sun-Yat-sen University 4School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications |
| Pseudocode | No | The paper describes the proposed architecture and components with diagrams and textual explanations, but does not include structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Code of this project is at https://github.com/chenx2000/DWMT. |
| Open Datasets | Yes | The simulation experiments are conducted on two datasets, i.e., CAVE (Park et al. 2007) and KAIST (Choi et al. 2017). ... we use CAVE as the training set and select ten scenes from KAIST for testing. |
| Dataset Splits | No | The paper mentions CAVE as the training set and KAIST for testing, but no explicit validation dataset split information is provided. |
| Hardware Specification | Yes | The DWMT is implemented in Py Torch and trained and tested on a single RTX A6000 GPU. |
| Software Dependencies | No | The paper states that the model is 'implemented in Py Torch', but no specific version number for PyTorch or any other software dependency is provided. |
| Experiment Setup | Yes | The model employs the Adam (Kingma and Ba 2015) optimizer (β1 = 0.9 and β2 = 0.999) with 500 epochs and an initial learning rate of 4 10 4. The learning rate is adjusted by Cosine Annealing scheme. The batch size is set to 5, and the training objective is to minimize the Root Mean Square Error (RMSE) between the reconstructed image and the corresponding ground truth. |