{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,9]],"date-time":"2026-05-09T09:00:35Z","timestamp":1778317235695,"version":"3.51.4"},"reference-count":31,"publisher":"MDPI AG","issue":"5","license":[{"start":{"date-parts":[[2024,2,24]],"date-time":"2024-02-24T00:00:00Z","timestamp":1708732800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Science Foundation","award":["1853864"],"award-info":[{"award-number":["1853864"]}]},{"name":"National Science Foundation","award":["2120943"],"award-info":[{"award-number":["2120943"]}]},{"name":"National Science Foundation","award":["2230034"],"award-info":[{"award-number":["2230034"]}]},{"name":"National Science Foundation","award":["2230035"],"award-info":[{"award-number":["2230035"]}]},{"name":"National Science Foundation","award":["2033521"],"award-info":[{"award-number":["2033521"]}]},{"name":"Google.org\u2019s Impact Challenge for Climate Innovation Program","award":["1853864"],"award-info":[{"award-number":["1853864"]}]},{"name":"Google.org\u2019s Impact Challenge for Climate Innovation Program","award":["2120943"],"award-info":[{"award-number":["2120943"]}]},{"name":"Google.org\u2019s Impact Challenge for Climate Innovation Program","award":["2230034"],"award-info":[{"award-number":["2230034"]}]},{"name":"Google.org\u2019s Impact Challenge for Climate Innovation Program","award":["2230035"],"award-info":[{"award-number":["2230035"]}]},{"name":"Google.org\u2019s Impact Challenge for Climate Innovation Program","award":["2033521"],"award-info":[{"award-number":["2033521"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>This paper assesses trending AI foundation models, especially emerging computer vision foundation models and their performance in natural landscape feature segmentation. While the term foundation model has quickly garnered interest from the geospatial domain, its definition remains vague. Hence, this paper will first introduce AI foundation models and their defining characteristics. Built upon the tremendous success achieved by Large Language Models (LLMs) as the foundation models for language tasks, this paper discusses the challenges of building foundation models for geospatial artificial intelligence (GeoAI) vision tasks. To evaluate the performance of large AI vision models, especially Meta\u2019s Segment Anything Model (SAM), we implemented different instance segmentation pipelines that minimize the changes to SAM to leverage its power as a foundation model. A series of prompt strategies were developed to test SAM\u2019s performance regarding its theoretical upper bound of predictive accuracy, zero-shot performance, and domain adaptability through fine-tuning. The analysis used two permafrost feature datasets, ice-wedge polygons and retrogressive thaw slumps because (1) these landform features are more challenging to segment than man-made features due to their complicated formation mechanisms, diverse forms, and vague boundaries; (2) their presence and changes are important indicators for Arctic warming and climate change. The results show that although promising, SAM still has room for improvement to support AI-augmented terrain mapping. The spatial and domain generalizability of this finding is further validated using a more general dataset EuroCrops for agricultural field mapping. Finally, we discuss future research directions that strengthen SAM\u2019s applicability in challenging geospatial domains.<\/jats:p>","DOI":"10.3390\/rs16050797","type":"journal-article","created":{"date-parts":[[2024,2,26]],"date-time":"2024-02-26T10:40:17Z","timestamp":1708944017000},"page":"797","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":29,"title":["Segment Anything Model Can Not Segment Anything: Assessing AI Foundation Model\u2019s Generalizability in Permafrost Mapping"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2237-9499","authenticated-orcid":false,"given":"Wenwen","family":"Li","sequence":"first","affiliation":[{"name":"School of Geographical Sciences and Urban Planning, Arizona State University, Tempe, AZ 85287, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8923-1213","authenticated-orcid":false,"given":"Chia-Yu","family":"Hsu","sequence":"additional","affiliation":[{"name":"School of Geographical Sciences and Urban Planning, Arizona State University, Tempe, AZ 85287, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sizhe","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Geographical Sciences and Urban Planning, Arizona State University, Tempe, AZ 85287, USA"},{"name":"School of Computing and Augmented Intelligence, Arizona State University, Tempe, AZ 85287, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yezhou","family":"Yang","sequence":"additional","affiliation":[{"name":"School of Computing and Augmented Intelligence, Arizona State University, Tempe, AZ 85287, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hyunho","family":"Lee","sequence":"additional","affiliation":[{"name":"School of Geographical Sciences and Urban Planning, Arizona State University, Tempe, AZ 85287, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anna","family":"Liljedahl","sequence":"additional","affiliation":[{"name":"Woodwell Climate Research Center, Falmouth, MA 02540, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chandi","family":"Witharana","sequence":"additional","affiliation":[{"name":"Department of Natural Resources and the Environment, University of Connecticut, Storrs, CT 06269, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yili","family":"Yang","sequence":"additional","affiliation":[{"name":"Woodwell Climate Research Center, Falmouth, MA 02540, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6711-8466","authenticated-orcid":false,"given":"Brendan M.","family":"Rogers","sequence":"additional","affiliation":[{"name":"Woodwell Climate Research Center, Falmouth, MA 02540, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Samantha T.","family":"Arundel","sequence":"additional","affiliation":[{"name":"U.S. Geological Survey, Center of Excellence for Geospatial Information Science, Rolla, MO 65401, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Matthew B.","family":"Jones","sequence":"additional","affiliation":[{"name":"National Center for Ecological Analysis & Synthesis, University of California, Santa Barbara, CA 93106, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kenton","family":"McHenry","sequence":"additional","affiliation":[{"name":"National Center for Supercomputing Applications (NCSA), University of Illinois at Urbana, Champaign, IL 61820, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Patricia","family":"Solis","sequence":"additional","affiliation":[{"name":"School of Geographical Sciences and Urban Planning, Arizona State University, Tempe, AZ 85287, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2024,2,24]]},"reference":[{"key":"ref_1","unstructured":"Devlin, J., Chang, M.W., Lee, K., and Toutanova, K. (2019, January 2\u20137). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2019, Minneapolis, MN, USA."},{"key":"ref_2","unstructured":"Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, \u0141., and Polosukhin, I. (2017). Attention is All You Need. Adv. Neural Inf. Process. Syst., 30."},{"key":"ref_3","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"181","DOI":"10.14358\/PERS.21-00059R2","article-title":"An Optimal GeoAI Workflow for Pan-Arctic Permafrost Feature Detection from High-Resolution Satellite Imagery","volume":"88","author":"Udawalpola","year":"2022","journal-title":"Photogramm. Eng. Remote Sens."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Li, W., Hsu, C.Y., Wang, S., Witharana, C., and Liljedahl, A. (2022, January 1). Real-time GeoAI for high-resolution mapping and segmentation of arctic permafrost features: The case of ice-wedge polygons. Proceedings of the 5th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery, Seattle, WA, USA.","DOI":"10.1145\/3557918.3565869"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"113495","DOI":"10.1016\/j.rse.2023.113495","article-title":"Mapping retrogressive thaw slumps using deep neural networks","volume":"288","author":"Yang","year":"2023","journal-title":"Remote Sens. Environ."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Li, Y., Wu, C.Y., Fan, H., Mangalam, K., Xiong, B., Malik, J., and Feichtenhofer, C. (2022, January 18\u201324). Mvitv2: Improved multiscale vision transformers for classification and detection. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.00476"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Li, W., and Hsu, C.Y. (2022). GeoAI for Large-Scale Image Analysis and Machine Vision: Recent Progress of Artificial Intelligence in Geography. ISPRS Int. J. Geo-Inf., 11.","DOI":"10.3390\/ijgi11070385"},{"key":"ref_9","first-page":"71","article-title":"GeoAI: Where Machine Learning and Big Data Converge in GIScience","volume":"20","author":"Li","year":"2020","journal-title":"J. Spat. Inf. Sci."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"e2015759118","DOI":"10.1073\/pnas.2015759118","article-title":"Replication across Space and Time Must Be Weak in the Social and Environmental Sciences","volume":"118","author":"Goodchild","year":"2021","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_11","unstructured":"Cha, K., Seo, J., and Lee, T. (2023). A Billion-scale Foundation Model for Remote Sensing Images. arXiv."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"4205","DOI":"10.1109\/JSTARS.2021.3070368","article-title":"On Creating Benchmark Dataset for Aerial Image Interpretation: Reviews, Guidances, and Million-Aid","volume":"14","author":"Long","year":"2021","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Lin, T.Y., Goyal, P., Girshick, R., He, K., and Doll\u00e1r, P. (2017, January 22\u201329). Focal Loss for Dense Object Detection. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref_14","first-page":"1","article-title":"Advancing Plain Vision Transformer towards Remote Sensing Foundation Model","volume":"61","author":"Wang","year":"2022","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_15","unstructured":"Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., and Clark, J. (2021, January 13). Learning Transferable Visual Models from Natural Language Supervision. Proceedings of the International Conference on Machine Learning. PMLR, Virtual."},{"key":"ref_16","unstructured":"Jiang, P. (2023, May 05). Conect Segment-Anything with CLIP. Available online: https:\/\/github.com\/PengtaoJiang\/Segment-Anything-CLIP."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., Xiao, T., Whitehead, S., Berg, A.C., and Lo, W.Y. (2023). Segment Anything. arXiv.","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"ref_18","unstructured":"Martin, D., Fowlkes, C., Tal, D., and Malik, J. (2001, January 7\u201314). A Database of Human Segmented Natural Images and its Application to Evaluating Segmentation Algorithms and Measuring Ecological Statistics. Proceedings of the 8th IEEE International Conference on Computer Vision, Vancouver, BC, Canada."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Li, J., Yang, T., Ji, W., Wang, J., and Cheng, L. (2022, January 18\u201324). Exploring denoised cross-video contrast for weakly-supervised temporal action localization. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.01929"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Gupta, A., Dollar, P., and Girshick, R. (2019, January 15\u201320). Lvis: A dataset for large vocabulary instance segmentation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00550"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Doll\u00e1r, P., and Zitnick, C.L. (2014, January 6\u201312). Microsoft coco: Common objects in context. Proceedings of the Computer Vision\u2013ECCV 2014: 13th European Conference, Zurich, Switzerland. Proceedings, Part V 13.","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"312","DOI":"10.1038\/ngeo2674","article-title":"Pan-Arctic Ice-Wedge Degradation in Warming Permafrost and Its Influence on Tundra Hydrology","volume":"9","author":"Liljedahl","year":"2016","journal-title":"Nat. Geosci."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"111297","DOI":"10.1016\/j.rse.2019.111297","article-title":"A raster version of the Circumpolar Arctic Vegetation Map (CAVM)","volume":"232","author":"Raynolds","year":"2019","journal-title":"Remote Sens. Environ."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Nitze, I., Heidler, K., Barth, S., and Grosse, G. (2021). Developing and testing a deep learning approach for mapping retrogressive thaw slumps. Remote Sens., 13.","DOI":"10.3390\/rs13214294"},{"key":"ref_25","unstructured":"Nitze, I., Heidler, K., Barth, S., Grosse, G., and Bernhard, P. (2022, January 12\u201316). Towards a Deep-Learning based Inventory of Retrogressive Thaw Slumps across the Arctic. Proceedings of the AGU Fall Meeting Abstracts, Chicago, IL, USA."},{"key":"ref_26","unstructured":"Meredith, M., Sommerkorn, M., Cassotta, S., Derksen, C., Ekaykin, A., Hollowed, A., Kofinas, G., Mackintosh, A., Melbourne-Thomas, J., and Muelbert, M. (2019). Polar Regions. Chapter 3, IPCC Special Report on the Ocean and Cryosphere in a Changing Climate."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"091001","DOI":"10.1088\/1748-9326\/ac8c5a","article-title":"Incorporating permafrost into climate mitigation and adaptation policy","volume":"17","author":"Natali","year":"2022","journal-title":"Environ. Res. Lett."},{"key":"ref_28","unstructured":"(2023, September 17). Avery. Our ML Tooling 2022\u2014Part 2: Data Operations. Available online: https:\/\/developmentseed.org\/blog\/2022-12-19-ml-tooling-2."},{"key":"ref_29","unstructured":"Jakubik, J., Chu, L., Fraccaro, P., Gomes, C., Nyirjesy, G., Bangalore, R., Lambhate, D., Das, K., Oliveira Borges, D., and Kimura, D. (2023). Prithvi-100M. arXiv."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"612","DOI":"10.1038\/s41597-023-02517-0","article-title":"EuroCrops: The Largest Harmonized Open Crop Dataset Across the European Union","volume":"10","author":"Schneider","year":"2023","journal-title":"Sci. Data"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"101715","DOI":"10.1016\/j.compenvurbsys.2021.101715","article-title":"GeoAI in Terrain Analysis: Enabling Multi-Source Deep Learning and Data Fusion for Natural Feature Detection","volume":"90","author":"Wang","year":"2021","journal-title":"Comput. Environ. Urban Syst."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/16\/5\/797\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T14:04:26Z","timestamp":1760105066000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/16\/5\/797"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,2,24]]},"references-count":31,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2024,3]]}},"alternative-id":["rs16050797"],"URL":"https:\/\/doi.org\/10.3390\/rs16050797","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,2,24]]}}}