Geographical Knowledge-Driven Representation Learning for Remote Sensing Images
By Wenyuan Li, Keyan Chen, Zhenwei Shi. 2021
95 citations recorded by the source corpus.
The proliferation of remote sensing satellites has resulted in a massive amount of remote sensing images. However, due to human and material resource constraints, the vast majority of remote sensing images remain unlabeled. As a result, it cannot be applied to currently available deep learning methods. To fully utilize the remaining unlabeled images, we propose a Geographical Knowledge-driven Representation (GeoKR) learning method for remote sensing images, improving network performance and reduce the demand for annotated data. The global land cover products and geographical location associated with each remote sensing image are regarded as geographical knowledge to provide supervision for representation learning and network pretraining. An efficient pretraining framework is proposed to eliminate the supervision noises caused by imaging times and resolutions difference between remote sensing images and geographical knowledge. A large-scale pretraining dataset Levir-KR is constructed to support network pretraining. It contains 1,431,950 remote sensing images from Gaofen series satellites with various resolutions. Experimental results demonstrate that our proposed method outperforms ImageNet pretraining and self-supervised representation learning methods and significantly reduces the burden of data annotation on downstream tasks, such as scene classification, semantic segmentation, object detection, and cloud/snow detection. It demonstrates that our proposed method can be used as a novel paradigm for pretraining neural networks. Codes will be available on https://github.com/flyakon/Geographical-Knowledge-driven-Representaion-Learning.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Media Technology
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
In Computer Vision and Pattern Recognition
In Media Technology
Citations in the corpus
- Continual Barlow Twins: Continual Self-Supervised Learning for Remote Sensing Semantic Segmentation (2022, methodology): “GeoKR [38] uses metadata for an efficient pretraining strategy”
- RingMo: A Remote Sensing Foundation Model With Masked Image Modeling (2023, methodology): “Furthermore, some researchers [24], [25], [29], [30] introduce additional information specific to RS (geographic information, time-series data, audio data, etc.) through careful search or matching and encourage models to learn similarities with additional information in addition to data augmentation.”
- SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery (2022, background): “[45] leverages geographical knowledge to aid SSL, which, however, can be difficult to obtain as annotations.”
- Self-Supervised Learning in Remote Sensing: A review (2022, background): “Li et al. [162] made use of GlobeLand30 [163], a global land cover product that divides the earth into different areas and includes ten different land cover types.”
- RSPrompter: Learning to Prompt for Remote Sensing Instance Segmentation Based on Visual Foundation Model (2023, background): “Despite the inherent big data attributes of remote sensing since its inception [11], [34], there has yet to be the development of foundational models specifically tailored to this field.”
- Advancing Plain Vision Transformer Toward Remote Sensing Foundation Model (2022, background): “[31] leverages the land cover ratio in the digital geographical information product as a weak label for RS representation learning.”
- RemoteCLIP: A Vision Language Foundation Model for Remote Sensing (2023, background): “Li et al. [36] distilled geographical vegetation for the same purpose.”