Corpus ID: 220380967

AutoAssign: Differentiable Label Assignment for Dense Object Detection

  title={AutoAssign: Differentiable Label Assignment for Dense Object Detection},
  author={Benjin Zhu and J. Wang and Zhengkai Jiang and Fuhang Zong and Songtao Liu and Zeming Li and J. Sun},
  • Benjin Zhu, J. Wang, +4 authors J. Sun
  • Published 2020
  • Computer Science
  • ArXiv
  • In this paper, we propose an anchor-free object detector with a fully differentiable label assignment strategy, named AutoAssign. It automatically determines positive/negative samples by generating positive and negative weight maps to modify each location's prediction dynamically. Specifically, we present a center weighting module to adjust the category-specific prior distributions and a confidence weighting module to adapt the specific assign strategy of each instance. The entire label… CONTINUE READING
    1 Citations

    Figures, Tables, and Topics from this paper.

    FCOS: A simple and strong anchor-free object detector


    Learning From Noisy Anchors for One-Stage Object Detection
    • 4
    • Highly Influential
    • PDF
    FreeAnchor: Learning to Match Anchors for Visual Object Detection
    • 42
    • Highly Influential
    • PDF
    CenterNet: Keypoint Triplets for Object Detection
    • 227
    • PDF
    You Only Look Once: Unified, Real-Time Object Detection
    • 9,367
    • Highly Influential
    • PDF
    Objects365: A Large-Scale, High-Quality Dataset for Object Detection
    • Shuai Shao, Zeming Li, +5 authors Jian Sun
    • Computer Science
    • 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
    • 2019
    • 19
    • PDF
    DenseBox: Unifying Landmark Localization with End to End Object Detection
    • 195
    • PDF
    SSD: Single Shot MultiBox Detector
    • 8,812
    • PDF
    Focal Loss for Dense Object Detection
    • 2,822
    • Highly Influential
    FCOS: Fully Convolutional One-Stage Object Detection
    • 339
    • Highly Influential
    • PDF
    RepPoints: Point Set Representation for Object Detection
    • 87
    • PDF