Deep Graph Neural Networks for Multi-Image Super Resolution Reconstruction
Keywords:
Multi-Image Super-Resolution, Deep Graph Neural Networks, Image Reconstruction, Inter-Image Relationships, Cross-Image DependenciesAbstract
This research introduces a pioneering approach to Multi-Image Super-Resolution (MISR) reconstruction, leveraging the power of Deep Graph Neural Networks (GNNs). Recognizing the limitations of traditional single-image super-resolution methods, the proposed framework exploits the inter-image relationships within a set of low-resolution images. A graph-based representation is introduced, where nodes correspond to image patches, and edges capture spatial correlations. Deep GNNs are integrated into this graph structure to facilitate information exchange and refinement of high-resolution estimations. This novel approach enables the model to exploit cross-image dependencies, resulting in a more holistic understanding of the scene and enhanced reconstruction. Through an iterative learning process, this method effectively leverages the contextual information from neighboring patches within and across multiple input images. Extensive experiments across the BSD100 dataset demonstrate the superior performance of the proposed MISR reconstruction method, showcasing remarkable improvements in image quality and finer details. The incorporation of Deep GNNs in the multi-image context proves to be a promising avenue for advancing the state-of-the-art in super-resolution reconstruction, offering a robust solution for applications requiring high-quality image enhancement.
Downloads
References
Ran, R., Deng, L. J., Jiang, T. X., Hu, J. F., Chanussot, J., & Vivone, G. (2023). GuidedNet: A general CNN fusion framework via high-resolution guidance for hyperspectral image super-resolution. IEEE Transactions on Cybernetics.
Lepcha, D. C., Goyal, B., Dogra, A., & Goyal, V. (2023). Image super-resolution: A comprehensive review, recent trends, challenges and applications. Information Fusion, 91, 230-260.
Wang, Z., Zhang, L., Shuai, T., Liang, S., & Li, S. (2022). Image Super-Resolution Reconstruction Based on Dual Residual Network.
Liu, M., Gao, H., & Ji, S. (2020, August). Towards deeper graph neural networks. In Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining (pp. 338-348).
Hui, Z., Wang, X., & Gao, X. (2018). Fast and accurate single image super-resolution via information distillation network. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 723-731).
Ahn, N., Kang, B., & Sohn, K. A. (2022). Efficient deep neural network for photo-realistic image super-resolution. Pattern Recognition, 127, 108649.
Mei, Y., Fan, Y., & Zhou, Y. (2021). Image Super-Resolution with Non-Local Sparse Attention. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 3517-3526,
Wu, J., Yue, T., Shen, Q., Cao, X., & Ma, Z. (2017). Multiple-Image Super Resolution Using Both Reconstruction Optimization and Deep Neural Network. IEEE Global Conference on Signal and Information Processing (GlobalSIP), 1175-1179.
Li, J., Lv, Q., Zhang, W., Zhu, B., Zhang, G., & Tan, Z. (2023). Multi-Attention Multi-Image Super-Resolution Transformer (MAST) for Remote Sensing. Remote Sensing, 15(17), 4183.
Tarasiewicz, T., Nalepa, J., & Kawulok, M. (2021). A Graph Neural Network For Multiple-Image Super-Resolution. IEEE International Conference on Image Processing (ICIP), 1824-1828.
Liu, F., Yang, X., & De Baets, B. (2023). A deep recursive multi-scale feature fusion network for image super-resolution. Journal of Visual Communication and Image Representation, 90, 103730.
Khaledyan, D., Amirany, A., Jafari, K., Moaiyeri, M. H., Khuzani, A. Z., & Mashhadi, N. (2020, October). Low-cost implementation of bilinear and bicubic image interpolation for real-time image super-resolution. In 2020 IEEE Global Humanitarian Technology Conference (GHTC) (pp. 1-5). IEEE.
R. Timofte, V. De Smet, and L. Van Gool. A+: Adjusted anchored neighborhood regression for fast super-resolution. In ACCV, 2014.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
