This repository is a curated collection of academic papers focused on semi-supervised and supervised learning techniques for sea ice concentration and sea ice type classification. It emphasizes research on methodologies related to both concentration estimation and type classification, including approaches for patch-level image classification and pixel-level semantic segmentation, providing a comprehensive resource for researchers and practitioners in the field.
Sea ice classification is a crucial task for understanding and predicting Arctic and Antarctic ice conditions. This repository aggregates key papers that explore various machine learning techniques, including semi-supervised and supervised learning approaches, to classify sea ice type and concentration effectively. The focus is on methodologies that target both patch-level classification and pixel-level semantic segmentation.
Paper Title | Year |
---|---|
Weakly Supervised Learning for Pixel-Level Sea Ice Concentration Extraction Using AI4Arctic Sea Ice Challenge Dataset | 2023 |
We welcome contributions to this repository! Please see our Contribution Guide for more details.
This project is licensed under the MIT License - see the LICENSE file for details.