Publication Type
Conference Proceeding Article
Version
acceptedVersion
Publication Date
12-2024
Abstract
Remote sensing (RS) imagery, requiring specialized satellites to collect and being difficult to annotate, suffers from data scarcity and class imbalance in certain spectrums. Due to data scarcity, training any large-scale RS models from scratch is unrealistic, and the alternative is to transfer pre-trained models by fine-tuning or a more data-efficient method LoRA. Due to class imbalance, transferred models exhibit strong bias, where features of the major class dominate over those of the minor class. In this paper, we propose debLoRA---a generic training approach that works with any LoRA variants to yield debiased features. It is an unsupervised learning approach that can diversify minor class features based on the shared attributes with major classes, where the attributes are obtained by a simple step of clustering.
Keywords
SAR, Remote-sensing images, LoRA, Visual transformers
Discipline
Categorical Data Analysis | Databases and Information Systems
Research Areas
Data Science and Engineering; Information Systems and Management
Publication
Proceedings of the 38th Conference on Neural Information Processing Systems, Vancouver, Canada, 2024 December 10-15
Publisher
The Neural Information Processing Systems Foundation
City or Country
California
Citation
TIAN, Zichen; CHEN, Zhaozheng; and SUN, Qianru.
Learning de-biased representations for remote-sensing imagery. (2024). Proceedings of the 38th Conference on Neural Information Processing Systems, Vancouver, Canada, 2024 December 10-15.
Available at: https://ink.library.smu.edu.sg/sis_research/9400
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.