本文介绍了一个在空间和尺度上全活跃特征交互(fully active feature interaction across both space and scales)的特征金字塔transformer模型,简称FPT。该模型将transformer和Feature Pyramid结合,可用于像素级的任务,在论文中作者进行了目标检测和实力分割,都取得了比较好的效果。为了讲解清楚,若有transformer不懂的读者,关于transformer可以在公众号中看另一篇文《Transformer解读》
One of the great advantages of using a deep learning framework to build recommender models is the freedom to build rich, flexible feature representations.