Segment object with few examples by generating pseudo-episodes from unlabeled data with IPE
Segment object with few examples by generating pseudo-episodes from unlabeled data with IPE
Image to Pseudo-Episode: Boosting Few-Shot Segmentation by Unlabeled Data
arXiv paper abstract https://arxiv.org/abs/2405.08765
arXiv PDF paper https://arxiv.org/pdf/2405.08765
Few-shot segmentation (FSS) aims to train a model which can segment the object from novel classes with a few labeled samples.
… Considering that there are abundant unlabeled data available, it is promising to improve the generalization ability by exploiting these various data.
For leveraging unlabeled data, … propose a novel method, named Image to Pseudo-Episode (IPE), to generate pseudo-episodes from unlabeled data.
… method contains two modules, i.e., the pseudo-label generation module and the episode generation module.
The former module generates pseudo-labels from unlabeled images by the spectral clustering algorithm, and the latter module generates pseudo-episodes from pseudo-labeled images by data augmentation methods.
… method achieves the state-of-the-art performance for FSS.
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website