Feature Pyramid Networks for Object Detection
key words: object detection, multi-scale feature map
文章摘抄：
- Feature pyramids are a basic component in recognition systems for detecting objects at different scales. Recent deep learning object detectors have avoided pyramid representationsns, in part becasuse they are compute and memory intesnsive.
- In this paper, we exploit the inherent multi-scale, pyramidal hierarchy of deep convolutional networks to construct feature pyramids with marginal extra cost.
- A top down architecture with lateral connections is developed for building high-level semantic feature maps at all scales.