Corpus ID: 211069177

CTM: Collaborative Temporal Modeling for Action Recognition

@article{Liu2020CTMCT,
  title={CTM: Collaborative Temporal Modeling for Action Recognition},
  author={Li-Yu Daisy Liu and Tao Wang and Jianbin Liu and Yang Guan and Qi Bu and Longfei Yang},
  journal={ArXiv},
  year={2020},
  volume={abs/2002.03152}
}
  • Li-Yu Daisy Liu, Tao Wang, +3 authors Longfei Yang
  • Published 2020
  • Computer Science
  • ArXiv
  • With the rapid development of digital multimedia, video understanding has become an important field. For action recognition, temporal dimension plays an important role, and this is quite different from image recognition. In order to learn powerful feature of videos, we propose a Collaborative Temporal Modeling (CTM) block (Figure 1) to learn temporal information for action recognition. Besides a parameter-free identity shortcut, as a separate temporal modeling block, CTM includes two… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 23 REFERENCES

    Temporal Segment Networks: Towards Good Practices for Deep Action Recognition

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    STM: SpatioTemporal and Motion Encoding for Action Recognition

    VIEW 1 EXCERPT

    D3D: Distilled 3D Networks for Video Action Recognition

    Grouped Spatial-Temporal Aggregation for Efficient Action Recognition

    VIEW 1 EXCERPT

    A Closer Look at Spatiotemporal Convolutions for Action Recognition

    VIEW 2 EXCERPTS

    Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset

    VIEW 1 EXCERPT

    Learning Spatio-Temporal Representation With Local and Global Diffusion