Lightweight recurrent neural network for image super-resolution

Abstract

In recent years, significant progress has been made in image super-resolution through the use of large-scale models. However, the efficacy of these models comes at the cost of their substantial size, posing challenges and limitations when deploying them on resource-constrained devices. Despite their remarkable performance, the feasibility of employing such models on low-end devices has remained a contentious topic. In light of this, our research introduces a lightweight approach to image super-resolution, leveraging a simple recurrent neural network architecture consisting of a recurrent convolution block. Our proposed model uses less than 75k parameters, which is 10 times fewer than the state-of-the-art transformer-based super-resolution model. Despite its small size, the proposed model performs well in image super-resolution tasks both visually and quantitatively. Our work presents a promising direction for …

Cite

@INPROCEEDINGS{10647844,
  author={Hossain, Mir Sazzat and Rahman, Akm Mahbubur and Amin, Md. Ashraful and Ali, Amin Ahsan},
  booktitle={2024 IEEE International Conference on Image Processing (ICIP)},
  title={Lightweight Recurrent Neural Network for Image Super-Resolution},
  year={2024},
  volume={},
  number={},
  pages={1567-1573},
  keywords={Performance evaluation;Recurrent neural networks;Costs;Convolution;Computational modeling;Superresolution;Transformers;Single Image Super-Resolution;Recurrent Neural Networks;Efficient Super-Resolution},
  doi={10.1109/ICIP51287.2024.10647844}
}