Bounding Box Regression With Uncertainty for Accurate Object Detection

@article{He2019BoundingBR,
  title={Bounding Box Regression With Uncertainty for Accurate Object Detection},
  author={Yihui He and Chenchen Zhu and Jianren Wang and Marios Savvides and Xiangyu Zhang},
  journal={2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2019},
  pages={2883-2892}
}
  • Yihui He, Chenchen Zhu, +2 authors Xiangyu Zhang
  • Published 2019
  • Computer Science
  • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • Large-scale object detection datasets (e.g., MS-COCO) try to define the ground truth bounding boxes as clear as possible. [...] Key Method The learned localization variance allows us to merge neighboring bounding boxes during non-maximum suppression (NMS), which further improves the localization performance. On MS-COCO, we boost the Average Precision (AP) of VGG-16 Faster R-CNN from 23.6% to 29.1%. More importantly, for ResNet-50-FPN Mask R-CNN, our method improves the AP and AP90 by 1.8% and 6.2% respectively…Expand Abstract

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 61 CITATIONS, ESTIMATED 90% COVERAGE

    Side-Aware Boundary Localization for More Precise Object Detection

    VIEW 1 EXCERPT
    CITES BACKGROUND

    KL-Divergence-Based Region Proposal Network for Object Detection

    VIEW 4 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Overlap Sampler for Region-Based Object Detection

    VIEW 1 EXCERPT
    CITES BACKGROUND

    FILTER CITATIONS BY YEAR

    2018
    2020

    CITATION STATISTICS

    • 7 Highly Influenced Citations

    • Averaged 20 Citations per year from 2018 through 2020

    • 23% Increase in citations per year in 2020 over 2019

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 75 REFERENCES

    Mask R-CNN

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

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

    VIEW 1 EXCERPT

    Focal Loss for Dense Object Detection

    VIEW 1 EXCERPT

    Object Detection via a Multi-region and Semantic Segmentation-Aware CNN Model

    SSD: Single Shot MultiBox Detector

    VIEW 1 EXCERPT

    Cascade R-CNN: Delving Into High Quality Object Detection

    VIEW 3 EXCERPTS