Corpus ID: 222208633

Deformable DETR: Deformable Transformers for End-to-End Object Detection

  title={Deformable DETR: Deformable Transformers for End-to-End Object Detection},
  author={X. Zhu and Weijie Su and Lewei Lu and Bin Li and X. Wang and Jifeng Dai},
  • X. Zhu, Weijie Su, +3 authors Jifeng Dai
  • Published 2020
  • Computer Science
  • ArXiv
  • DETR has been recently proposed to eliminate the need for many hand-designed components in object detection while demonstrating good performance. However, it suffers from slow convergence and limited feature spatial resolution, due to the limitation of Transformer attention modules in processing image feature maps. To mitigate these issues, we proposed Deformable DETR, whose attention modules only attend to a small set of key sampling points around a reference. Deformable DETR can achieve… CONTINUE READING
    1 Citations

    Figures and Tables from this paper.

    Sparse R-CNN: End-to-End Object Detection with Learnable Proposals


    Deformable Convolutional Networks
    • 1,247
    • PDF
    FCOS: Fully Convolutional One-Stage Object Detection
    • 343
    • PDF
    Focal Loss for Dense Object Detection
    • 2,843
    • Highly Influential
    Deformable ConvNets V2: More Deformable, Better Results
    • 240
    • PDF
    Deep Learning for Generic Object Detection: A Survey
    • 361
    • Highly Influential
    • PDF
    M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network
    • 134
    • PDF
    Feature Pyramid Networks for Object Detection
    • 4,509
    • Highly Influential
    • PDF
    Deep Feature Pyramid Reconfiguration for Object Detection
    • 66
    • Highly Influential
    • PDF
    NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection
    • 197
    • PDF
    RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
    • 13
    • PDF