An improved object detection algorithm based on depthwise separable convolutions

@inproceedings{Yu2020AnIO,
  title={An improved object detection algorithm based on depthwise separable convolutions},
  author={Xiu-yuan Yu and Qiliang L. Bao and Haolong Jia and Yu Li and Rui Qin},
  booktitle={Target Recognition and Artificial Intelligence Summit Forum},
  year={2020}
}
  • Xiu-yuan Yu, Qiliang L. Bao, +2 authors Rui Qin
  • Published in
    Target Recognition and…
    2020
  • Engineering, Computer Science
  • Aiming at small objects detection such as unmanned aerial vehicle (UAV), this paper proposes a fast object detection algorithm based on depth wise separable convolutions. Firstly, the inverted residuals units based on depth wise convolutions and pointwise convolutions are used to construct a lightweight feature extraction network to improve the network’s speed. Secondly, the feature pyramid network is used to detect the five scale feature maps to improve the detection performance of small… CONTINUE READING

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

    References

    Publications referenced by this paper.
    SHOWING 1-8 OF 8 REFERENCES

    Feature Pyramid Networks for Object Detection

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    You Only Look Once: Unified, Real-Time Object Detection

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    SSD: Single Shot MultiBox Detector[C].European

    • W Liu, D Anguelov, D Erhan
    • Conference on Computer Vision. Springer International Publishing,
    • 2016
    VIEW 2 EXCERPTS

    Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

    VIEW 1 EXCERPT

    Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation

    VIEW 3 EXCERPTS