Corpus ID: 220042190

Self-Convolution: A Highly-Efficient Operator for Non-Local Image Restoration

@article{Guo2020SelfConvolutionAH,
  title={Self-Convolution: A Highly-Efficient Operator for Non-Local Image Restoration},
  author={Lanqing Guo and S. Ravishankar and Bihan Wen},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.13714}
}
  • Lanqing Guo, S. Ravishankar, Bihan Wen
  • Published 2020
  • Computer Science
  • ArXiv
  • Constructing effective image priors is critical to solving ill-posed inverse problems, such as image reconstruction. Recent works proposed to exploit image non-local similarity for inverse problems by grouping similar patches, and demonstrated state-of-the-art results in many applications. However, comparing to classic local methods based on filtering or sparsity, most of the non-local algorithms are timeconsuming, mainly due to the highly inefficient and redundant block matching step, where… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 42 REFERENCES

    Image Recovery via Transform Learning and Low-Rank Modeling: The Power of Complementary Regularizers

    VIEW 6 EXCERPTS

    Multi-spectral SIFT for scene category recognition

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    When sparsity meets low-rankness: Transform learning with non-local low-rank constraint for image restoration

    VIEW 8 EXCERPTS

    Cross-Field Joint Image Restoration via Scale Map

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    FFDNet: Toward a Fast and Flexible Solution for CNN-Based Image Denoising

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    FOCNet: A Fractional Optimal Control Network for Image Denoising

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Flickr1024: A Large-Scale Dataset for Stereo Image Super-Resolution

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    IEEE transactions on pattern analysis and machine intelligence

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL