# A Scale-Arbitrary Image Super-Resolution Network Using Frequency-domain Information

arXiv: https://arxiv.org/abs/2212.04314 · HTML: https://mlsanity.com/paper/2212.04314/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 54 references that are in the map.
- First posted on arXiv 8 Dec 2022.

## Builds on (most cited first, 24 of 24)

- [Asynchronous Methods for Deep Reinforcement Learning](https://mlsanity.com/paper/1602.01783/) (2016)
- [Meta-SR: A Magnification-Arbitrary Network for Super-Resolution](https://mlsanity.com/paper/1903.00875/) (2019)
- [Deep Learning for Single Image Super-Resolution: A Brief Review](https://mlsanity.com/paper/1808.03344/) (2018)
- [Local Texture Estimator for Implicit Representation Function](https://mlsanity.com/paper/2111.08918/) (2021)
- [Details or Artifacts: A Locally Discriminative Learning Approach to Realistic Image Super-Resolution](https://mlsanity.com/paper/2203.09195/) (2022)
- [Image and Video Compression with Neural Networks: A Review](https://mlsanity.com/paper/1904.03567/) (2019)
- [Fourier Space Losses for Efficient Perceptual Image Super-Resolution](https://mlsanity.com/paper/2106.00783/) (2021)
- [Discrete Cosine Transform Network for Guided Depth Map Super-Resolution](https://mlsanity.com/paper/2104.06977/) (2021)
- [Learning A Single Network for Scale-Arbitrary Super-Resolution](https://mlsanity.com/paper/2004.03791/) (2020)
- [Wide Activation for Efficient and Accurate Image Super-Resolution](https://mlsanity.com/paper/1808.08718/) (2018)
- [Task-Driven Super Resolution: Object Detection in Low-resolution Images](https://mlsanity.com/paper/1803.11316/) (2018)
- [DAQ: Channel-Wise Distribution-Aware Quantization for Deep Image Super-Resolution Networks](https://mlsanity.com/paper/2012.11230/) (2020)
- [SRWarp: Generalized Image Super-Resolution under Arbitrary Transformation](https://mlsanity.com/paper/2104.10325/) (2021)
- [Fine Perceptive GANs for Brain MR Image Super-Resolution in Wavelet Domain](https://mlsanity.com/paper/2011.04145/) (2020)
- [A Two-Stage Attentive Network for Single Image Super-Resolution](https://mlsanity.com/paper/2104.10488/) (2021)
- [MDCN: Multi-scale Dense Cross Network for Image Super-Resolution](https://mlsanity.com/paper/2008.13084/) (2020)
- [Towards Bidirectional Arbitrary Image Rescaling: Joint Optimization and Cycle Idempotence](https://mlsanity.com/paper/2203.00911/) (2022)
- [OverNet: Lightweight Multi-Scale Super-Resolution with Overscaling Network](https://mlsanity.com/paper/2008.02382/) (2020)
- [A Frequency Domain Neural Network for Fast Image Super-resolution](https://mlsanity.com/paper/1712.03037/) (2017)
- [FreqNet: A Frequency-domain Image Super-Resolution Network with Dicrete Cosine Transform](https://mlsanity.com/paper/2111.10800/) (2021)
- [Multi-grained Attention Networks for Single Image Super-Resolution](https://mlsanity.com/paper/1909.11937/) (2019)
- [Cross-SRN: Structure-Preserving Super-Resolution Network with Cross Convolution](https://mlsanity.com/paper/2201.01458/) (2022)
- [Adaptive Transform Domain Image Super-resolution Via Orthogonally Regularized Deep Networks](https://mlsanity.com/paper/1904.10082/) (2019)
- [Improving image classifiers for small datasets by learning rate adaptations](https://mlsanity.com/paper/1903.10726/) (2019)

## Abstract (arXiv)

Image super-resolution (SR) is a technique to recover lost high-frequency information in low-resolution (LR) images. Spatial-domain information has been widely exploited to implement image SR, so a new trend is to involve frequency-domain information in SR tasks. Besides, image SR is typically application-oriented and various computer vision tasks call for image arbitrary magnification. Therefore, in this paper, we study image features in the frequency domain to design a novel scale-arbitrary image SR network. First, we statistically analyze LR-HR image pairs of several datasets under different scale factors and find that the high-frequency spectra of different images under different scale factors suffer from different degrees of degradation, but the valid low-frequency spectra tend to be retained within a certain distribution range. Then, based on this finding, we devise an adaptive scale-aware feature division mechanism using deep reinforcement learning, which can accurately and adaptively divide the frequency spectrum into the low-frequency part to be retained and the high-frequency one to be recovered. Finally, we design a scale-aware feature recovery module to capture and fuse multi-level features for reconstructing the high-frequency spectrum at arbitrary scale factors. Extensive experiments on public datasets show the superiority of our method compared with state-of-the-art methods.