• Published 2019

) 2 3 n Category score Shape offset confidence score Shape offset Category score Shape offset Single Shot Module ( SSM ) ( One stage ) ( Two stage )

@inproceedings{Li201923,
  title={) 2 3 n Category score Shape offset confidence score Shape offset Category score Shape offset Single Shot Module ( SSM ) ( One stage ) ( Two stage )},
  author={Yuezun Li and Xian Bian and Ming-Ching Chang and Siwei Lyu},
  year={2019}
}
Recent works succeeded to generate adversarial perturbations on the entire image or the object of interests to corrupt CNN based object detectors. In this paper, we focus on exploring the vulnerability of the Single Shot Module (SSM) commonly used in recent object detectors, by adding small perturbations to patches in the background outside the object. The SSM is referred to the Region Proposal Network used in a two-stage object detector or the single-stage object detector itself. The SSM is… CONTINUE READING

Figures and Tables from this paper.

References

Publications referenced by this paper.
SHOWING 1-10 OF 33 REFERENCES

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

VIEW 8 EXCERPTS
HIGHLY INFLUENTIAL

YOLOv3: An Incremental Improvement

VIEW 5 EXCERPTS
HIGHLY INFLUENTIAL

SSD: Single Shot MultiBox Detector

VIEW 7 EXCERPTS
HIGHLY INFLUENTIAL

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

VIEW 7 EXCERPTS
HIGHLY INFLUENTIAL

Robust Physical Adversarial Attack on Faster R-CNN Object Detector

VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL

FSSD: Feature Fusion Single Shot Multibox Detector

VIEW 4 EXCERPTS
HIGHLY INFLUENTIAL

DPATCH: An Adversarial Patch Attack on Object Detectors

VIEW 1 EXCERPT