BoT Net

文章目录

• 1. Motivation

  • 2. Contribution

  • 3. Method

  • • 3.1 Architecture Network
    • 3.2 MHSA
    • 3.3 Relative Position Encoding

  • 4. Experiment

  • • 4.1 BoTNet improves over ResNet on COCO In- stance Segmentation with Mask R-CNN
    • 4.2. Scale Jitter helps BoTNet more than ResNet
    • 4. 3 Relative Position Encodings BoostPerformance BoTNet
    • 4.4 Why replace all three c5 spatial convolutions?
    • 4.5 BoTNet improves backbones in ResNet Family