Mode Seeking Generative Adversarial Networks for Diverse Image Synthesis

@article{Mao2019ModeSG,
  title={Mode Seeking Generative Adversarial Networks for Diverse Image Synthesis},
  author={Qi Mao and Hsin-Ying Lee and Hung-Yu Tseng and Siwei Ma and Ming-Hsuan Yang},
  journal={2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2019},
  pages={1429-1437}
}
  • Qi Mao, Hsin-Ying Lee, +2 authors Ming-Hsuan Yang
  • Published 2019
  • Computer Science
  • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • Most conditional generation tasks expect diverse outputs given a single conditional context. However, conditional generative adversarial networks (cGANs) often focus on the prior conditional information and ignore the input noise vectors, which contribute to the output variations. Recent attempts to resolve the mode collapse issue for cGANs are usually task-specific and computationally expensive. In this work, we propose a simple yet effective regularization term to address the mode collapse… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 36 CITATIONS, ESTIMATED 90% COVERAGE

    Hierarchical Modes Exploring in Generative Adversarial Networks

    VIEW 5 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Multimodal Image Outpainting with Regularized Normalized Diversification

    VIEW 8 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    On the Evaluation of Conditional GANs

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Stochastic Conditional Generative Networks with Basis Decomposition

    VIEW 12 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    When Relation Networks meet GANs: Relation GANs with Triplet Loss

    VIEW 2 EXCERPTS

    A Simple yet Effective Way for Improving the Performance of GANs

    VIEW 3 EXCERPTS
    CITES BACKGROUND & METHODS

    FILTER CITATIONS BY YEAR

    2018
    2020

    CITATION STATISTICS

    • 9 Highly Influenced Citations

    • Averaged 12 Citations per year from 2018 through 2020

    • 92% Increase in citations per year in 2020 over 2019

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 43 REFERENCES

    High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs

    VIEW 1 EXCERPT

    StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networks

    VIEW 9 EXCERPTS
    HIGHLY INFLUENTIAL

    Multi-agent Diverse Generative Adversarial Networks

    VIEW 2 EXCERPTS

    Image-to-Image Translation with Conditional Adversarial Networks

    Generative Multi-Adversarial Networks

    VIEW 2 EXCERPTS