Corpus ID: 214641150

Data Uncertainty Learning in Face Recognition

@article{Chang2020DataUL,
  title={Data Uncertainty Learning in Face Recognition},
  author={Jie Chang and Zhonghao Lan and Changmao Cheng and Yichen Wei},
  journal={ArXiv},
  year={2020},
  volume={abs/2003.11339}
}
  • Jie Chang, Zhonghao Lan, +1 author Yichen Wei
  • Published 2020
  • Computer Science
  • ArXiv
  • Modeling data uncertainty is important for noisy images, but seldom explored for face recognition. The pioneer work, PFE, considers uncertainty by modeling each face image embedding as a Gaussian distribution. It is quite effective. However, it uses fixed feature (mean of the Gaussian) from an existing model. It only estimates the variance and relies on an ad-hoc and costly metric. Thus, it is not easy to use. It is unclear how uncertainty affects feature learning. This work applies data… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 53 REFERENCES

    Probabilistic Face Embeddings

    VIEW 8 EXCERPTS
    HIGHLY INFLUENTIAL

    Robust Person Re-Identification by Modelling Feature Uncertainty

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    A Light CNN for Deep Face Representation With Noisy Labels

    VIEW 1 EXCERPT

    On the Capacity of Face Representation

    VIEW 1 EXCERPT

    The Devil of Face Recognition is in the Noise

    VIEW 1 EXCERPT

    Noise-Tolerant Paradigm for Training Face Recognition CNNs

    VIEW 1 EXCERPT

    A data-driven approach to cleaning large face datasets

    VIEW 1 EXCERPT