26 citations · 57 across the 21 of their papers we have counts for
23 papers
Beyond Accuracy: Assessing Calibration of Geospatial Foundation Models and Their Sensitivity to Distribution Shifts
Nils Lehmann, Jakob Gawlikowski, Burak Ekim +2
Geospatial Foundation Models (GeoFMs) are most commonly ranked and selected by accuracy on standard benchmark conditions via averaged ranks. We show that this protocol is too narro…
Earth Embeddings
Adam J. Stewart, Heng Fang, Isaac A. Corley +1
Earth observation is moving from foundation models that users must run themselves toward embedding products that package model feature outputs as reusable data without needing to d…
RSGPNet: Geometric Prompting for Remote Sensing Open-Vocabulary Semantic Segmentation
Shanwen Wang, Xin Sun, Sirui Wang +1
Open-vocabulary semantic segmentation (OVSS) enables text-guided segmentation of unseen objects, breaking fixed-class limitations to achieve open-world understanding. However, exis…
Agentic AI for Remote Sensing: Technical Challenges and Research Directions
Muhammad Akhtar Munir, Muhammad Umer Sheikh, Akashah Shabbir +5
Earth Observation (EO) is moving beyond static prediction toward multi-step analytical workflows that require coordinated reasoning over data, tools, and geospatial state. While fo…
SpectralEarth-FM: Bringing Hyperspectral Imagery into Multimodal Earth Observation Pretraining
Nassim Ait Ali Braham, Aaron Banze, Conrad M. Albrecht +3
Earth observation (EO) foundation models (FMs) are increasingly trained on multisensor data, spanning multispectral imagery (MSI), synthetic aperture radar (SAR), and derived geosp…
Bias-Constrained Diffusion Schedules for PDE Emulations: Reconstruction Error Minimization and Efficient Unrolled Training
Constantin Le Cleï, Nils Thuerey, Xiaoxiang Zhu
Conditional Diffusion Models are powerful surrogates for emulating complex spatiotemporal dynamics, yet they often fail to match the accuracy of deterministic neural emulators for…
Earth Embeddings as Products: Taxonomy, Ecosystem, and Standardized Access
Heng Fang, Adam J. Stewart, Isaac Corley +2
Geospatial Foundation Models (GFMs) provide powerful representations, but high compute costs hinder their widespread use. Pre-computed embedding data products offer a practical "fr…
OpenEarthAgent: A Unified Framework for Tool-Augmented Geospatial Agents
Akashah Shabbir, Muhammad Umer Sheikh, Muhammad Akhtar Munir +8
Recent progress in multimodal reasoning has enabled agents that interpret imagery, connect it with language, and execute structured analytical tasks. Extending these capabilities t…
Unified Primitive Proxies for Structured Shape Completion
Zhaiyu Chen, Yuqing Wang, Xiao Xiang Zhu
Structured shape completion recovers missing geometry as primitives rather than as unstructured points, which enables primitive-based surface reconstruction. Instead of following t…
TSE-Net: Semi-supervised Monocular Height Estimation from Single Remote Sensing Images
Sining Chen, Xiao Xiang Zhu
Monocular height estimation plays a critical role in 3D perception for remote sensing, offering a cost-effective alternative to multi-view or LiDAR-based methods. While deep learni…
GEWDiff: Geometric Enhanced Wavelet-based Diffusion Model for Hyperspectral Image Super-resolution
Sirui Wang, Jiang He, Natàlia Blasco Andreo +1
Improving the quality of hyperspectral images (HSIs), such as through super-resolution, is a crucial research area. However, generative modeling for HSIs presents several challenge…
Physically consistent and uncertainty-aware learning of spatiotemporal dynamics
Qingsong Xu, Jonathan L Bamber, Nils Thuerey +5
Accurate long-term forecasting of spatiotemporal dynamics remains a fundamental challenge across scientific and engineering domains. Existing machine learning methods often neglect…
Learning Generalizable Shape Completion with SIM(3) Equivariance
Yuqing Wang, Zhaiyu Chen, Xiao Xiang Zhu
3D shape completion methods typically assume scans are pre-aligned to a canonical frame. This leaks pose and scale cues that networks may exploit to memorize absolute positions rat…
UrbanSAM: Learning Invariance-Inspired Adapters for Segment Anything Models in Urban Construction
Chenyu Li, Danfeng Hong, Bing Zhang +4
Object extraction and segmentation from remote sensing (RS) images is a critical yet challenging task in urban environment monitoring. Urban morphology is inherently complex, with…
REO-VLM: Transforming VLM to Meet Regression Challenges in Earth Observation
Xizhe Xue, Guoting Wei, Hao Chen +4
The rapid evolution of Vision Language Models (VLMs) has catalyzed significant advancements in artificial intelligence, expanding research across various disciplines, including Ear…
Physics-embedded Fourier Neural Network for Partial Differential Equations
Qingsong Xu, Nils Thuerey, Yilei Shi +3
We consider solving complex spatiotemporal dynamical systems governed by partial differential equations (PDEs) using frequency domain-based discrete learning approaches, such as Fo…
Causal Graph Neural Networks for Wildfire Danger Prediction
Shan Zhao, Ioannis Prapas, Ilektra Karasante +4
Wildfire forecasting is notoriously hard due to the complex interplay of different factors such as weather conditions, vegetation types and human activities. Deep learning models s…
Physics-aware Machine Learning Revolutionizes Scientific Paradigm for Machine Learning and Process-based Hydrology
Qingsong Xu, Yilei Shi, Jonathan Bamber +3
Accurate hydrological understanding and water cycle prediction are crucial for addressing scientific and societal challenges associated with the management of water resources, part…
Towards Large-scale Building Attribute Mapping using Crowdsourced Images: Scene Text Recognition on Flickr and Problems to be Solved
Yao Sun, Anna Kruspe, Liqiu Meng +4
Crowdsourced platforms provide huge amounts of street-view images that contain valuable building information. This work addresses the challenges in applying Scene Text Recognition…
SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal Dataset for Self-Supervised Learning in Earth Observation
Yi Wang, Nassim Ait Ali Braham, Zhitong Xiong +3
Self-supervised pre-training bears potential to generate expressive representations without human annotation. Most pre-training in Earth observation (EO) are based on ImageNet or m…
Large-scale Building Height Retrieval from Single SAR Imagery based on Bounding Box Regression Networks
Yao Sun, Lichao Mou, Yuanyuan Wang +2
Building height retrieval from synthetic aperture radar (SAR) imagery is of great importance for urban applications, yet highly challenging owing to the complexity of SAR data. Thi…
SurroundNet: Towards Effective Low-Light Image Enhancement
Fei Zhou, Xin Sun, Junyu Dong +2
Although Convolution Neural Networks (CNNs) has made substantial progress in the low-light image enhancement task, one critical problem of CNNs is the paradox of model complexity a…
Changes in Twitter geolocations: Insights and suggestions for future usage
Anna Kruspe, Matthias Häberle, Eike J. Hoffmann +3
Twitter data has become established as a valuable source of data for various application scenarios in the past years. For many such applications, it is necessary to know where Twit…