Yin-Yin He, Peizhen Zhang, Xiu-Shen Wei, Xiangyu Zhang, Jian Sun
TL;DR本研究探讨使用混淆矩阵对长尾实例分割问题中不同类别间的精细误分类信息进行建模,从而解决训练样本不平衡引起的模型偏差问题,提出的 Pairwise Class Balance (PCB) 方法能够有效地进行模型规范化训练,实验结果表明该方法具有较强的泛化性能和优越的表现。
Abstract
long-tailed instance segmentation is a challenging task due to the extreme imbalance of training samples among classes. It causes severe biases of the head classes (with majority samples) against the tailed ones. This renders "how to appropriately define and alleviate the bias" one of