Corpus ID: 53041849

ScratchDet: Exploring to Train Single-Shot Object Detectors from Scratch

@article{Zhu2018ScratchDetET,
  title={ScratchDet: Exploring to Train Single-Shot Object Detectors from Scratch},
  author={Rui Zhu and Shifeng Zhang and Xiaobo Wang and Longyin Wen and Hailin Shi and Liefeng Bo and Tao Mei},
  journal={ArXiv},
  year={2018},
  volume={abs/1810.08425}
}
  • Rui Zhu, Shifeng Zhang, +4 authors Tao Mei
  • Published 2018
  • Computer Science
  • ArXiv
  • Current state-of-the-art object objectors are fine-tuned from the off-the-shelf networks pretrained on large-scale classification dataset ImageNet, which incurs some additional problems: 1) The classification and detection have different degrees of sensitivity to translation, resulting in the learning objective bias; 2) The architecture is limited by the classification network, leading to the inconvenience of modification. To cope with these problems, training detectors from scratch is a… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 16 CITATIONS

    Propose-and-Attend Single Shot Detector

    VIEW 2 EXCERPTS
    CITES METHODS

    Multi-Scale Receptive Field Detection Network

    VIEW 2 EXCERPTS
    CITES METHODS

    VisDrone-DET2018: The Vision Meets Drone Object Detection in Image Challenge Results

    VIEW 1 EXCERPT
    CITES METHODS

    Multi-Scale Receptive Field Detection Network FIRST

    VIEW 2 EXCERPTS
    CITES METHODS

    Understanding the Effects of Pre-Training for Object Detectors via Eigenspectrum

    VIEW 8 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Recent Advances in Deep Learning for Object Detection

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Improved Selective Refinement Network for Face Detection

    VIEW 2 EXCERPTS
    CITES BACKGROUND

    Object Detection in 20 Years: A Survey

    VIEW 1 EXCERPT
    CITES BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 44 REFERENCES

    DSOD: Learning Deeply Supervised Object Detectors from Scratch

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Single-Shot Refinement Neural Network for Object Detection

    VIEW 6 EXCERPTS

    DetNet: A Backbone network for Object Detection

    VIEW 1 EXCERPT

    Focal Loss for Dense Object Detection

    VIEW 3 EXCERPTS
    HIGHLY INFLUENTIAL

    RON: Reverse Connection with Objectness Prior Networks for Object Detection

    Training Region-Based Object Detectors with Online Hard Example Mining

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

    VIEW 1 EXCERPT

    R-FCN: Object Detection via Region-based Fully Convolutional Networks

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

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

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL