Nov, 2022
有礼貌的教师:互相学习和伪标签阈值的半监督实例分割
Polite Teacher: Semi-Supervised Instance Segmentation with Mutual
Learning and Pseudo-Label Thresholding
TL;DRPolite Teacher使用相互学习框架,结合有限的监督信息和无标定实例,通过confidence thresholding和mask scoring来过滤噪声,进而在anchor-free detector上提高了8%的mask AP值,这是关于半监督实例分割问题的首个anchor-free解决方案。