Corpus ID: 215238650

Temporally Distributed Networks for Fast Video Semantic Segmentation

@article{Hu2020TemporallyDN,
  title={Temporally Distributed Networks for Fast Video Semantic Segmentation},
  author={Ping Hu and Fabian Caba Heilbron and Oliver Wang and Zhe Lin and Stan Sclaroff and Federico Perazzi},
  journal={ArXiv},
  year={2020},
  volume={abs/2004.01800}
}
  • Ping Hu, Fabian Caba Heilbron, +3 authors Federico Perazzi
  • Published 2020
  • Computer Science, Engineering
  • ArXiv
  • We present TDNet, a temporally distributed network designed for fast and accurate video semantic segmentation. We observe that features extracted from a certain high-level layer of a deep CNN can be approximated by composing features extracted from several shallower sub-networks. Leveraging the inherent temporal continuity in videos, we distribute these sub-networks over sequential frames. Therefore, at each time step, we only need to perform a lightweight computation to extract a sub-features… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    Citations

    Publications citing this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 68 REFERENCES

    Low-Latency Video Semantic Segmentation

    VIEW 12 EXCERPTS
    HIGHLY INFLUENTIAL

    Semantic Video Segmentation by Gated Recurrent Flow Propagation

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Accel: A Corrective Fusion Network for Efficient Semantic Segmentation on Video

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Budget-Aware Deep Semantic Video Segmentation

    VIEW 3 EXCERPTS

    Architecture Search of Dynamic Cells for Semantic Video Segmentation

    VIEW 1 EXCERPT

    Efficient Video Semantic Segmentation with Labels Propagation and Refinement

    VIEW 1 EXCERPT

    Dynamic Video Segmentation Network

    STD2P: RGBD Semantic Segmentation Using Spatio-Temporal Data-Driven Pooling

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Semantic Video CNNs Through Representation Warping

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Clockwork Convnets for Video Semantic Segmentation

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL