LGPConv: Learnable Gaussian Perturbation Convolution for Lightweight Pansharpening

Authors: Chen-Yu Zhao, Tian-Jing Zhang, Ran Ran, Zhi-Xuan Chen, Liang-Jian Deng

IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We incorporate LGPConv into a well-designed pansharpening network and demonstrate its superiority through extensive experiments, achieving state-of-the-art performance with minimal parameters (27K). and 4 Experiments.
Researcher Affiliation Academia University of Electronic Science and Technology of China, Chengdu, 611731 chenyuzhaouestc@gmail.com, zhangtianjinguestc@163.com, {ranran, 2019050305012}@std.uestc.edu.cn, liangjian.deng@uestc.edu.cn
Pseudocode No The paper does not include any structured pseudocode or algorithm blocks.
Open Source Code Yes Code is available on the Git Hub page of the authors.
Open Datasets Yes Due to the page limitation, dataset, metrics, compared methods, and training platform are mentioned in supplementary. ... quantitative results of different methods on the 1258 samples from WV-3 dataset with eight bands. ... assess the proposed method on 4-band datasets, i.e., GF-2 and QB data.
Dataset Splits No The paper mentions total sample numbers for the datasets used (e.g., 1258 samples from WV-3 dataset, 81 samples from GF-2 dataset and 48 samples from QB dataset) but does not explicitly specify the training, validation, and test splits (e.g., percentages or exact sample counts for each split) in the main text.
Hardware Specification No The paper states Due to the page limitation, dataset, metrics, compared methods, and training platform are mentioned in supplementary, indicating that hardware specifications are not provided in the main text.
Software Dependencies No The paper states Due to the page limitation, dataset, metrics, compared methods, and training platform are mentioned in supplementary, and does not provide specific software dependencies with version numbers in the main text.
Experiment Setup Yes The number of Resnet-like blocks in LGPConv-Net. Since we have set the number of Resnet-like blocks (N) to 4 in default, the influence of Resnet-like blocks will be demonstrated by justifying its number. We have tested the performance of LGPConv-Net with the different number of Resnet-like blocks in the deep inference block, where N = 2, 4, 6, 8.