Semi-supervised semantic segmentation via multi-level and multi-view perturbation consistency

Semi-supervised semantic segmentation (SSS) aims to achieve fine-grained pixel-level image annotations by using a small set of labeled training data and a large amount of unlabeled data. Typical SSS methods exploit various types of augmentations to introduce weak-to-strong perturbation consistency o...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiacheng Yang, Hui Fang, Jing Dong, Rui Liu, Pengfei Yi, Jiajun Yin, Qiang Zhang
Format: Default Article
Published: 2025
Subjects:
Information and computing sciences
Artificial intelligence
Information systems
Applied mathematics
Numerical and computational mathematics
Semi-supervised semantic segmentation
Perturbation consistency
Augmentation
Semi-supervised learning
Online Access:https://hdl.handle.net/2134/30764915.v1
Tags: Add Tag
No Tags, Be the first to tag this record!

Similar Items