Active Learning for Segmentation Based on Bayesian Sample Queries
We release code and demo scripts for our work on querying new samples to maximize representativeness within a dataset. Within the scope of semantic segmentation, our proposed approach vastly reduces the amount of necessary annotated samples to achieve performance on par with having the full dataset being annotated.
If you find this work useful, please cite using the following:
@article{ozdemir2021active,
title = {Active learning for segmentation based on Bayesian sample queries},
journal = {Knowledge-Based Systems},
volume = {214},
pages = {106531},
year = {2021},
issn = {0950-7051},
doi = {https://doi.org/10.1016/j.knosys.2020.106531},
url = {https://www.sciencedirect.com/science/article/pii/S0950705120306602},
author = {Firat Ozdemir and Zixuan Peng and Philipp Fuernstahl and Christine Tanner and Orcun Goksel},
}
