As large-scale heterogeneous data sets become increasingly available, adapting foundation models at low cost has become a key issue. Seminal works in natural language processing, e.g. Low-Rank Adaptation (LoRA), leverage the low "intrinsic rank" of parameter updates during adaptation. In this paper, we argue that incorporating stronger inductive biases in both data and models can enhance the adaptation of Geospatial Foundation Models (GFMs), pretrained on RGB satellite images, to other types of optical satellite data. Specifically, the pretrained parameters of GFMs serve as a strong prior for the spatial structure of multispectral images. For this reason, we introduce DEFLECT (Deflecting Embeddings for Finetuning Latent representations for Earth and Climate Tasks), a novel strategy for adapting GFMs to multispectral satellite imagery with very few additional parameters. DEFLECT improves the representation capabilities of the extracted features, particularly enhancing spectral information, which is essential for geoscience and environmental-related tasks. We demonstrate the effectiveness of our method across three different GFMs and five diverse datasets, ranging from forest monitoring to marine environment segmentation. Compared to competing methods, DEFLECT achieves on-par or higher accuracy with 5-10$\times$ fewer parameters for classification and segmentation tasks. The code will be made publicly available.
Machine learning in remote sensing has matured alongside a proliferation in availability and resolution of geospatial imagery, but its utility is bottlenecked by the need for labeled data. What's more, many labeled geospatial datasets are specific to certain regions, instruments, or extreme weather events. We investigate the application of modern domain-adaptation to multiple proposed geospatial benchmarks, uncovering unique challenges and proposing solutions to them.
Source-Free Domain Adaptation for Geospatial Point Cloud Semantic Segmentation
Aerospace Information Research Institute, Chinese Academy of Sciences（中国科学院航天信息研究所）；International Research Center of Big Data for Sustainable Development Goals（可持续发展目标大数据国际研究中心）；University of Chinese Academy of Sciences（中国科学院大学）；Zhengzhou Institute for Advanced Research of Henan Polytechnic University（河南理工大学郑州研究院）；Henan Polytechnic University（河南理工大学）；School of Aeronautic Engineering, Changsha University of Science and Technology（长沙理工大学航空工程学院）；China University of Geosciences, Beijing（中国地质大学（北京））
A Sensor Agnostic Domain Generalization Framework for Leveraging Geospatial Foundation Models: Enhancing Semantic Segmentation viaSynergistic Pseudo-Labeling and Generative Learning
University of Houston（休斯敦大学）；Purdue University（普渡大学）
Survey of Multimodal Geospatial Foundation Models: Techniques, Applications, and Challenges
School of Artificial Intelligence and Robotics, Hunan University（湖南大学人工智能与机器人学院）；Institute of Remote Sensing and Geographic Information System, Peking University（北京大学遥感与地理信息系统研究所）；School of Automation, Central South University（中南大学自动化学院）；Electronic Information School, Wuhan University（武汉大学电子信息学院）；Helmholtz-Zentrum Dresden-Rossendorf（德累斯顿-罗斯托克亥姆霍尔茨中心）；Lancaster Environment Centre, Lancaster University（兰卡斯特大学环境研究中心）；Hyperspectral Computing Laboratory, Department of Technology of Computers and Communications, Escuela Politécnica, University of Extremadura（埃斯特雷马杜拉大学技术计算机与通讯系超光谱计算实验室）
Geospatial Foundation Models to Enable Progress on Sustainable Development Goals
Helmholtz-Zentrum Dresden-Rossendorf（德累斯顿-罗斯托克研究所）；University of Iceland（冰岛大学）；Wuhan University（武汉大学）；Wuhan University of Science and Technology（武汉科技大学）；Universitat de València（瓦伦西亚大学）
Adapting Prithvi-EO for Fallow Detection for Food-Water Nexus: ViT-Adapter Necks and Parameter-Efficient Backbone tuning of Geospatial Foundation Model
Earth, Atmospheric and Geospatial Science, Saint Louis University（圣路易斯大学地球、大气与地理空间科学系）
CrossEarth: Geospatial Vision Foundation Model for Domain Generalizable Remote Sensing Semantic Segmentation
School of Computer Science, Shanghai Jiao Tong University（上海交通大学计算机科学学院）；Key Laboratory of Cryospheric Science and Frozen Soil Engineering, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences（中国科学院寒区旱区环境与工程研究所冻土科学与冻土工程重点实验室）；School of Automation and Intelligent Sensing, Shanghai Jiao Tong University（上海交通大学自动化与智能感知学院）；School of Artificial Intelligence, Shanghai Jiao Tong University（上海交通大学人工智能学院）；School of Electronics and Communication Engineering, Sun Yat-sen University（中山大学电子与通信工程学院）；School of Computer Science, Wuhan University（武汉大学计算机科学学院）；Zhongguancun Academy（中关村学院）；State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University（武汉大学测绘遥感信息工程国家重点实验室）；School of Engineering Science, University of Science and Technology of China（中国科学技术大学工程科学学院）；School of Information and Electronics, Beijing Institute of Technology（北京理工大学信息与电子学院）
Leveraging AI multimodal geospatial foundation models for improved near-real-time flood mapping at a global scale
North Carolina State University（北卡罗来纳州立大学）；LabCorp（实验室集团）；Humboldt-Universität zu Berlin（柏林洪堡大学）；Earth and Life Institute, UCLouvain（鲁汶天主教大学地球与生命研究所）；University of Helsinki（赫尔辛基大学）；Clark University（克拉克大学）；Thünen Institute of Farm Economics（蒂嫩农场经济研究所）
