Cross-Loss Pseudo Labeling for Semi-Supervised Segmentation

Training semantic segmentation models requires pixel-level annotations, leading to a significant labeling cost in dataset creation. To alleviate this issue, recent research has focused on semi-supervised learning, which utilizes only a small amount of annotation. In this context, pseudo labeling tec...

Full description

Bibliographic Details

Published in:	IEEE Access
Main Authors:	Seungyeol Lee, Taeho Kim, Jae-Pil Heo
Format:	Article
Language:	English
Published:	IEEE 2023-01-01
Subjects:	Semi-supervised semantic segmentation confirmation bias pseudo-labeling
Online Access:	https://ieeexplore.ieee.org/document/10239380/