Please use this identifier to cite or link to this item: http://bura.brunel.ac.uk/handle/2438/27839
Full metadata record
DC FieldValueLanguage
dc.contributor.authorGala, J-
dc.contributor.authorNag, S-
dc.contributor.authorHuang, H-
dc.contributor.authorLiu, R-
dc.contributor.authorZhu, X-
dc.date.accessioned2023-12-10T17:52:49Z-
dc.date.available2023-12-10T17:52:49Z-
dc.date.issued2023-12-10-
dc.identifier.citationGala, J. et al. (2023) 'Adaptive-Labeling for Enhancing Remote Sensing Cloud Understanding', NeurIPS 2023 Workshop on Tackling Climate Change with Machine Learning: Blending New and Existing Knowledge Systems Workshop, New Orleans, LA, USA, 16 December, pp. 1 - 8. Available at: https://www.climatechange.ai/papers/neurips2023/4 (accessed: 10 December 2023 ).en_US
dc.identifier.urihttps://bura.brunel.ac.uk/handle/2438/27839-
dc.descriptionCode availability: https://github.com/Surrey-UPLab/Adaptive-Cloud-Label/tree/main .-
dc.description.abstractCloud analysis is a critical component of weather and climate science, impacting various sectors like disaster management. However, achieving fine-grained cloud analysis, such as cloud segmentation, in remote sensing remains challenging due to the inherent difficulties in obtaining accurate labels, leading to significant labeling errors in training data. Existing methods often assume the availability of reliable segmentation annotations, limiting their overall performance. To address this inherent limitation, we introduce an innovative model-agnostic Cloud Adaptive-Labeling (CAL) approach, which operates iteratively to enhance the quality of training data annotations and consequently improve the performance of the learned model. Our methodology commences by training a cloud segmentation model using the original annotations. Subsequently, it introduces a trainable pixel intensity threshold for adaptively labeling the cloud training images on-the-fly. The newly generated labels are then employed to fine-tune the model. Extensive experiments conducted on multiple standard cloud segmentation benchmarks demonstrate the effectiveness of our approach in significantly boosting the performance of existing segmentation models. Our CAL method establishes new state-of-the-art results when compared to a wide array of existing alternatives.en_US
dc.format.extent1 - 8-
dc.format.mediumElectronic-
dc.publisherUnited Nationsen_US
dc.relation.urihttps://www.climatechange.ai/papers/neurips2023/4-
dc.relation.urihttps://neurips.cc/virtual/2023/workshop/66543-
dc.relation.urihttps://github.com/Surrey-UPLab/Adaptive-Cloud-Label/tree/main-
dc.sourceNeural Information Processing Systems (NeurIPS) Workshop on Tackling Climate Change with Machine Learning-
dc.sourceNeural Information Processing Systems (NeurIPS) Workshop on Tackling Climate Change with Machine Learning-
dc.titleAdaptive-Labeling for Enhancing Remote Sensing Cloud Understandingen_US
dc.typeConference Paperen_US
pubs.publication-statusPublished online-
pubs.publisher-urlhttps://www.climatechange.ai/papers/neurips2023/4-
Appears in Collections:Dept of Economics and Finance Research Papers

Files in This Item:
File Description SizeFormat 
FullText.pdf519.02 kBAdobe PDFView/Open


Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.