Deep Stacked Hierarchical Multi-Patch Network for Image Deblurring

@article{Zhang2019DeepSH,
  title={Deep Stacked Hierarchical Multi-Patch Network for Image Deblurring},
  author={Hongguang Zhang and Yuchao Dai and Hongdong Li and Piotr Koniusz},
  journal={2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2019},
  pages={5971-5979}
}
  • Hongguang Zhang, Yuchao Dai, +1 author Piotr Koniusz
  • Published 2019
  • Computer Science
  • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • Despite deep end-to-end learning methods have shown their superiority in removing non-uniform motion blur, there still exist major challenges with the current multi-scale and scale-recurrent models: 1) Deconvolution/upsampling operations in the coarse-to-fine scheme result in expensive runtime; 2) Simply increasing the model depth with finer-scale levels cannot improve the quality of deblurring. To tackle the above problems, we present a deep {hierarchical multi-patch network} inspired by… CONTINUE READING

    Figures and Topics from this paper.

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 10 CITATIONS

    Improved Deep Multi-Patch Hierarchical Network With Nested Module for Dynamic Scene Deblurring

    VIEW 10 EXCERPTS
    CITES BACKGROUND, METHODS & RESULTS
    HIGHLY INFLUENCED

    Spatially-Attentive Patch-Hierarchical Network for Adaptive Motion Deblurring

    VIEW 11 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Scale-Iterative Upscaling Network for Image Deblurring

    VIEW 7 EXCERPTS
    CITES METHODS, BACKGROUND & RESULTS
    HIGHLY INFLUENCED

    Learning Event-Based Motion Deblurring

    VIEW 6 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    NTIRE 2020 Challenge on Image and Video Deblurring

    VIEW 4 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    NTIRE 2020 Challenge on Perceptual Extreme Super-Resolution: Methods and Results

    VIEW 1 EXCERPT
    CITES BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 32 REFERENCES

    Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurring

    Scale-Recurrent Network for Deep Image Deblurring

    VIEW 10 EXCERPTS
    HIGHLY INFLUENTIAL

    Dynamic Scene Deblurring Using Spatially Variant Recurrent Neural Networks

    VIEW 1 EXCERPT

    Deep Multi-patch Aggregation Network for Image Style, Aesthetics, and Quality Estimation

    VIEW 1 EXCERPT

    Generalized video deblurring for dynamic scenes

    Blur-Invariant Deep Learning for Blind-Deblurring

    VIEW 1 EXCERPT

    Deep Video Deblurring for Hand-Held Cameras

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks