Search Results

Documents authored by Vatsavai, Ranga Raju

Document
DOI: 10.4230/LIPIcs.GIScience.2025.18
U-Prithvi: Integrating a Foundation Model and U-Net for Enhanced Flood Inundation Mapping

Authors: Vit Kostejn, Yamil Essus, Jenna Abrahamson, and Ranga Raju Vatsavai

Published in: LIPIcs, Volume 346, 13th International Conference on Geographic Information Science (GIScience 2025)

Abstract
In recent years, large pre-trained models, commonly referred to as foundation models, have become increasingly popular for various tasks leveraging transfer learning. This trend has expanded to remote sensing, where transformer-based foundation models such as Prithvi, msGFM, and SatSwinMAE have been utilized for a range of applications. While these transformer-based models, particularly the Prithvi model, exhibit strong generalization capabilities, they have limitations on capturing fine-grained details compared to convolutional neural network architectures like U-Net in segmentation tasks. In this paper, we propose a novel architecture, U-Prithvi, which combines the strengths of the Prithvi transformer with those of U-Net. We introduce a RandomHalfMaskLayer to ensure balanced learning from both models during training. Our approach is evaluated on the Sen1Floods11 dataset for flood inundation mapping, and experimental results demonstrate better performance of U-Prithvi over both individual models, achieving improved performance on out-of-sample data. While this principle is illustrated using the Prithvi model, it is easily adaptable to other foundation models.
Cite as

Vit Kostejn, Yamil Essus, Jenna Abrahamson, and Ranga Raju Vatsavai. U-Prithvi: Integrating a Foundation Model and U-Net for Enhanced Flood Inundation Mapping. In 13th International Conference on Geographic Information Science (GIScience 2025). Leibniz International Proceedings in Informatics (LIPIcs), Volume 346, pp. 18:1-18:17, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025)

Copy BibTex To Clipboard

@InProceedings{kostejn_et_al:LIPIcs.GIScience.2025.18,
  author =	{Kostejn, Vit and Essus, Yamil and Abrahamson, Jenna and Vatsavai, Ranga Raju},
  title =	{{U-Prithvi: Integrating a Foundation Model and U-Net for Enhanced Flood Inundation Mapping}},
  booktitle =	{13th International Conference on Geographic Information Science (GIScience 2025)},
  pages =	{18:1--18:17},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-378-2},
  ISSN =	{1868-8969},
  year =	{2025},
  volume =	{346},
  editor =	{Sila-Nowicka, Katarzyna and Moore, Antoni and O'Sullivan, David and Adams, Benjamin and Gahegan, Mark},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.GIScience.2025.18},
  URN =		{urn:nbn:de:0030-drops-238479},
  doi =		{10.4230/LIPIcs.GIScience.2025.18},
  annote =	{Keywords: GeoAI, flood mapping, foundation model, U-Net, Prithvi}
}
Document
DOI: 10.4230/LIPIcs.GISCIENCE.2018.16
FUTURES-AMR: Towards an Adaptive Mesh Refinement Framework for Geosimulations

Authors: Ashwin Shashidharan, Ranga Raju Vatsavai, Derek B. Van Berkel, and Ross K. Meentemeyer

Published in: LIPIcs, Volume 114, 10th International Conference on Geographic Information Science (GIScience 2018)

Abstract
Adaptive Mesh Refinement (AMR) is a computational technique used to reduce the amount of computation and memory required in scientific simulations. Geosimulations are scientific simulations using geographic data, routinely used to predict outcomes of urbanization in urban studies. However, the lack of support for AMR techniques with geosimulations limits exploring prediction outcomes at multiple resolutions. In this paper, we propose an adaptive mesh refinement framework FUTURES-AMR, based on static user-defined policies to enable multi-resolution geosimulations. We develop a prototype for the cellular automaton based urban growth simulation FUTURES by exploiting static and dynamic mesh refinement techniques in conjunction with the Patch Growing Algorithm (PGA). While, the static refinement technique supports a statically defined fixed resolution mesh simulation at a location, the dynamic refinement technique supports dynamically refining the resolution based on simulation outcomes at runtime. Further, we develop two approaches - asynchronous AMR and synchronous AMR, suitable for parallel execution in a distributed computing environment with varying support for solution integration of the multi-resolution results. Finally, using the FUTURES-AMR framework with different policies in an urban study, we demonstrate reduced execution time, and low memory overhead for a multi-resolution simulation.
Cite as

Ashwin Shashidharan, Ranga Raju Vatsavai, Derek B. Van Berkel, and Ross K. Meentemeyer. FUTURES-AMR: Towards an Adaptive Mesh Refinement Framework for Geosimulations. In 10th International Conference on Geographic Information Science (GIScience 2018). Leibniz International Proceedings in Informatics (LIPIcs), Volume 114, pp. 16:1-16:15, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2018)

Copy BibTex To Clipboard

@InProceedings{shashidharan_et_al:LIPIcs.GISCIENCE.2018.16,
  author =	{Shashidharan, Ashwin and Vatsavai, Ranga Raju and Van Berkel, Derek B. and Meentemeyer, Ross K.},
  title =	{{FUTURES-AMR: Towards an Adaptive Mesh Refinement Framework for Geosimulations}},
  booktitle =	{10th International Conference on Geographic Information Science (GIScience 2018)},
  pages =	{16:1--16:15},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-083-5},
  ISSN =	{1868-8969},
  year =	{2018},
  volume =	{114},
  editor =	{Winter, Stephan and Griffin, Amy and Sester, Monika},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.GISCIENCE.2018.16},
  URN =		{urn:nbn:de:0030-drops-93440},
  doi =		{10.4230/LIPIcs.GISCIENCE.2018.16},
  annote =	{Keywords: Adaptive mesh refinement, Geosimulation, Distributed system, Multi-resolution, Urban geography}
}
Any Issues?
X

Feedback on the Current Page

CAPTCHA

Thanks for your feedback!

Feedback submitted to Dagstuhl Publishing

Could not send message

Please try again later or send an E-mail
© 2023-2025 Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact