Boosting VLAD with double assignment using deep features for action recognition in videos

@article{Duta2016BoostingVW,
  title={Boosting VLAD with double assignment using deep features for action recognition in videos},
  author={Ionut C. Duta and Tuan Anh Nguyen and Kiyoharu Aizawa and Bogdan Ionescu and Nicu Sebe},
  journal={2016 23rd International Conference on Pattern Recognition (ICPR)},
  year={2016},
  pages={2210-2215}
}
  • Ionut C. Duta, Tuan Anh Nguyen, +2 authors Nicu Sebe
  • Published in
    23rd International Conference…
    2016
  • Computer Science
  • The encoding method is an important factor for an action recognition pipeline. One of the key points for the encoding method is the assignment step. A very widely used super-vector encoding method is the vector of locally aggregated descriptors (VLAD), with very competitive results in many tasks. However, it considers only hard assignment and the criteria for the assignment is performed only from the features side, by looking for which visual word the features are voting. In this work we… CONTINUE READING

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    Citations

    Publications citing this paper.
    SHOWING 1-6 OF 6 CITATIONS

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 27 REFERENCES

    Recognizing 50 human action categories of web videos

    VIEW 9 EXCERPTS
    HIGHLY INFLUENTIAL

    Towards Good Practices for Very Deep Two-Stream ConvNets

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Action Recognition with Improved Trajectories

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning realistic human actions from movies

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Histograms of Motion Gradients for real-time video classification

    VIEW 1 EXCERPT

    A Robust and Efficient Video Representation for Action Recognition

    VIEW 2 EXCERPTS

    Beyond short snippets: Deep networks for video classification

    VIEW 2 EXCERPTS