The world is long-tailed. What does this mean for computer vision and visual
recognition? The main two implications are (1) the number of categories we need
to consider in applications can be very large, and (2) the number of training
examples for most categories can be very small. Current vi
在研究中,我们提出了 Decoupled Training for Devil in the Tails(DT2)的假设并开发了一个名为 Alternating Class-Balanced Sampling(ACBS)的新方法,以应对视觉关系模型的长尾分布问题。我们的结果表明,相比于更复杂的方法,DT2-ACBS 极大地提升了场景图生成任务的简单架构的性能。这提示在解决这个问题时需要同时考虑复杂模型的发展和长尾问题。