Direction-aware Spatial Context Features for Shadow Detection and Removal

@article{Hu2018DirectionawareSC,
  title={Direction-aware Spatial Context Features for Shadow Detection and Removal},
  author={Xiaowei Hu and Lei Zhu and Chi-Wing Fu and Jing Qin and Pheng-Ann Heng},
  journal={arXiv: Computer Vision and Pattern Recognition},
  year={2018}
}
  • Xiaowei Hu, Lei Zhu, +2 authors Pheng-Ann Heng
  • Published 2018
  • Computer Science
  • arXiv: Computer Vision and Pattern Recognition
  • Shadow detection and shadow removal are fundamental and challenging tasks, requiring an understanding of the global image semantics. This paper presents a novel deep neural network design for shadow detection and removal by analyzing the image context in a direction-aware manner. To achieve this, we first formulate the direction-aware attention mechanism in a spatial recurrent neural network (RNN) by introducing attention weights when aggregating spatial context features in the RNN. By learning… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 12 CITATIONS

    Bidirectional Feature Pyramid Network with Recurrent Attention Residual Modules for Shadow Detection

    VIEW 7 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Mask-ShadowGAN: Learning to Remove Shadows From Unpaired Data

    VIEW 16 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Shading-aware shadow detection and removal from a single image

    VIEW 9 EXCERPTS
    CITES METHODS, RESULTS & BACKGROUND
    HIGHLY INFLUENCED

    ARGAN: Attentive Recurrent Generative Adversarial Network for Shadow Detection and Removal

    VIEW 7 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Revisiting Shadow Detection: A New Benchmark Dataset for Complex World

    VIEW 6 EXCERPTS
    CITES BACKGROUND & METHODS

    Wavelet-Based Dual-Branch Network for Image Demoireing

    VIEW 7 EXCERPTS
    CITES METHODS
    HIGHLY INFLUENCED

    Conditional GANs for Multi-Illuminant Color Constancy: Revolution or yet Another Approach?

    • Oleksii Sidorov
    • Computer Science
    • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
    • 2019

    Direction-Aware Spatial Context Features for Shadow Detection

    VIEW 2 EXCERPTS
    CITES BACKGROUND & RESULTS

    Effective shadow removal via multi-scale image decomposition

    VIEW 3 EXCERPTS
    CITES BACKGROUND & METHODS

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 60 REFERENCES

    Learning to recognize shadows in monochromatic natural images

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Noisy Label Recovery for Shadow Detection in Unfamiliar Domains

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Paired Regions for Shadow Detection and Removal

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Pyramid Scene Parsing Network

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removal

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    DeshadowNet: A Multi-context Embedding Deep Network for Shadow Removal

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Direction-Aware Spatial Context Features for Shadow Detection

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL