BriefGPT.xyz
Mar, 2016
简单有效：弱监督实例和语义分割
Weakly Supervised Semantic Labelling and Instance Segmentation
HTML
PDF
Anna Khoreva, Rodrigo Benenson, Jan Hosang, Matthias Hein, Bernt Schiele
TL;DR
本文提出了一种无需修改分割训练过程的弱监督训练方法, 通过精心设计给定边界框的输入标签, 经过单一训练循环即可达到先前算法的弱监督结果并能够抵达完全监督模型的约95%的语义标注和实例分割质量。
Abstract
semantic labelling
and
instance segmentation
are two tasks that require particularly costly annotations. Starting from
weak supervision
in
→