Geospatial imagery plays an increasingly important role in environmental monitoring, climate resilience, agriculture, urban planning, disaster response, and humanitarian action. At the same time, imagery collected from satellites, aircraft, drones, and ground platforms presents fundamental computer vision challenges, including heterogeneous sensing modalities, large variations in scale and resolution, limited or noisy annotations, spatial and temporal domain shifts, and the need for reliable inference in open-world settings. GeoCV brings together the computer vision and geospatial imaging communities to advance methods for scalable, robust, and responsible analysis of Earth-observation data. The workshop provides a forum for researchers developing new computer vision and machine learning methods to engage directly with geospatial researchers addressing challenging real-world problems.
GeoCV 2027 builds on two successful prior editions at WACV. GeoCV 2025 was a full-day workshop featuring five keynote talks, 20 archival papers, nine extended abstracts, approximately 30 posters, and strong attendance throughout the day. GeoCV 2026 continued this momentum with three keynote talks, 16 papers, three extended abstracts, additional invited contributions, and once again drew a full room. The third edition will highlight emerging directions including multimodal geospatial foundation models, vision-language models, language-guided and agentic GeoAI, label-efficient learning, cross-sensor and cross-region generalization, multi-sensor fusion, and trustworthy geospatial AI, while continuing to emphasize impactful applications across Earth observation.
We invite authors to submit high-quality papers on computer vision and image analysis for geospatial imaging. Submitted manuscripts will be peer-reviewed, and refereed for originality, presentation, empirical results and overall quality. In addition to papers focused on algorithmic novelty, we also encourage papers that demonstrate effective deployment of recent computer vision paradigms to compelling geospatial imaging applications.
Topics of interest include (but are not limited to):
Geospatial foundation models, large vision models, and multimodal or vision-language models
Agentic and language-guided GeoAI, spatial reasoning, and interactive Earth-observation systems
Self-supervised, weakly supervised, few-shot, active, and other label-efficient learning
Domain adaptation and generalization, open-world recognition, out-of-distribution detection, and trustworthy geospatial AI
Multi-sensor and multi-temporal fusion across optical, SAR, LiDAR, thermal, and hyperspectral data
Generative modeling and computer vision methods including super-resolution, reconstruction, segmentation, detection, and change analysis
Applications in environmental and biodiversity monitoring, agriculture, climate, disaster response, urban systems, humanitarian applications, and sustainable development
Submission Types
Regular Research Papers
Regular papers should present original research and will undergo peer review. Papers must be 5–8 pages, excluding references, and follow the WACV 2027 formatting guidelines. Accepted regular papers that are presented at the workshop will be included in the WACV 2027 Workshop Proceedings.
Submissions will be handled through OpenReview.
Submission site for Regular Research Papers: Click here
Authors should follow the WACV 2027 formatting and submission requirements for regular workshop papers. Supplementary material may be included.
Saurabh Prasad
University of Houston
Jocelyn Chanussot
INRIA
Claudia Paris
University of Twente
Biplab Banerjee
Indian Institute of Technology, Bombay
Danfeng Hong
Southeast University
Chen Chen
University of Central Florida
Chen Chen is an Associate Professor of Computer Science at the University of Central Florida. He received his Ph.D. in Electrical Engineering from the University of Texas at Dallas in 2016, where he received the David Daniel Fellowship for Best Doctoral Dissertation. His research interests include computer vision, efficient deep learning, and federated learning, with applications to resource-aware machine vision, large-scale camera networks, and the Internet of Things.
Alina Zare
University of Florida
Alina Zare is a Professor of Electrical and Computer Engineering at the University of Florida and Director of the Artificial Intelligence and Informatics Research Institute. Her research focuses on machine learning and AI for automated understanding of complex data and imagery, with applications including hyperspectral imaging, remote sensing, plant phenotyping, sonar, LiDAR, ground-penetrating radar, and target and hazard detection.
Regular paper submission: October 14, October 16, 2026 AOE
Notification of acceptance: November 2, November 7 2026
Camera-ready deadline: November 20
GeoCV 2027 Workshop: January 4 or 5, 2027
Diagnose, Then Correct: Training-Free Post-Processing for Cross-View Geo-Localization - Yehong Jiang, Wilson Cao
Scaling Geospatial Foundation Models on the MajorTOM Dataset: CNNs, Transformers, and Mamba - A Case Study on the PhilEO Bench - Nikolaos Dionelis, Riccardo Musto, Jente Bosmans, Simone Sarti, Giancarlo Paoletti, Peter Jack Naylor, Valerio Marsocci, Sébastien Lefèvre, Bertrand Le Saux, Nicolas Longépé
Early-Fusion Cue Analysis for Thin-Obstacle Semantic Segmentation in UAV Navigation: A Systematic Empirical Study - Negar Fathi
SALT: State-Aligned Low-Rank Transfer for Single-Temporal Cross-Dataset Structural Localization in Remote Sensing - Aggelos Psiris, Georgios Angelidis, Giorgos Sfikas, Maria Vakalopoulou, Thomas Lagkas, Panagiotis Sarigiannidis, Georgios Th. Papadopoulos
Learning Annotation-Driven Latent Structure for Multi-Modal Fusion - Meilun Zhou, Aditya Dutt, Alina Zare
Satellite Pretraining Is Not One Thing: A Matched-Pair Study of Generic and Earth-Observation Foundation Models for Scene Classification - Sandarbh Agrawal, Rakshith Sathish, Ashwathi Nambiar
Data Efficient Image Synthesis For InfraRed Target Recognition via PCA Back-Projection - Surendar Devasundaram, Thierry Fournel, Vijayakumar Bhagavatula, Abhijit Mahalanobis,
Harvesting the Unseen: Active Cross-Modal Grounding for Fine-Grained Aerial Retrieval - Idan Kligvasser, Yotam Intrator, George Leifman, Bolous Jaber, Moshe Kimhi, Regev Cohen, Genady Beryozkin, Niv Efron, Ehud Rivlin,
Location-Conditioned Crash Risk Estimation from Satellite Imagery - Ahmad Elallaf, Noah Sizemore, Yu Zhang, Nathan Jacobs, Gongbo Liang
Distillation Meets Pruning: Compressing GeoFM for Optical-to-Radar Knowledge Transfer - Amina Said, Julia Dietlmeier, Margaret McCaul, Noel O'Connor
Multi-Method Attribution for Explainable Sensor Fusion in Sea Ice Concentration Retrieval - Muhammad Amir
MAPPED-IT: Non-Adversarial Diverse Hyperspectral-LiDAR Bidirectional Translation - Aditya Dutt, Alina Zare
Generalist vs. Specialist Vision Models for Building Attributes: A Cross-Region Evaluation with Time-Checked Labels - Seyedeh Mobina Noorani, Karla Saldana Ochoa
Forest Canopy Height Estimation from Harmonized Landsat and Sentinel-1 Observations Using Deep Learning and GEDI LiDAR - Steven Lu, Antonio Ferraz
Can Multimodal LLMs Predict Rapid Intensification? A Leakage-Controlled Zero-Shot Benchmark for Tropical Cyclones - Sukhada Ghewari, Harshal Tupsamudre
Right for the Right Reasons? Auditing Shortcut Learning in Methane Quantification from Sentinel-2 - Agata Wijata, Agata Sage, Kornel Bujak, Magdalena Chojnacka, Jakub Nalepa, Nicolas Longépé, Maciej Swiechowski, Przemyslaw Biecek
MoreGeo: Multi-Orientation Reference Embeddings for Robust Cross-View Geo-Localization - Neil Wang
Michael J. Bianco, Microsoft
Emanuele Dalsasso, Inria
Sudipan Saha, Indian Institute of Technology Delhi
Ushasi Chaudhuri, Tata Consultancy Services
Aditya Dutt, University of Florida
Ryan Faulkenberry, University of Houston
Anan Yaghmour, University of Houston
Takayuki Shinohara, National Institute of Advanced Industrial Science and Technology (AIST)
Chaman Banolia, Indian Institute of Technology Bombay
Constanza A. Molina Catricheo, University of Münster
Maram Hasan, Indian Institute of Technology Bombay
Mainak Singha, University of Trento
Gabriele Moser, University of Genoa
Negar Fathi, University of Nebraska-Lincoln
Kaveh Safavigerdini, University of Missouri
Md Aminur Hossain, Indian Institute of Technology Bombay
Amina Said, Dublin City University
Julia Dietlmeier, Insight Research Ireland Centre for Data Analytics
Agata Wijata, Silesian University of Technology
Judith Nkechinyere Njoku, University of Wyoming
Samrat Mukherjee, Indian Institute of Technology Bombay
Georgios Angelidis, Harokopio University
Aggelos Psiris, Harokopio University
Giorgos Sfikas, University of West Attica
Amey Sunil Kulkarni
Shivam Pande
Workshop program will be announced after conclusion of the peer-review period.
Past Workshops:
2nd Edition of GeoCV @ WACV 2026:
The program for the workshop is available here, and the workshop proceedings are available here: https://openaccess.thecvf.com/WACV2026_workshops/GeoCV
1st Edition of GeoCV @ WACV 2025:
The program for the workshop is available here, and the workshop proceedings are available here: https://openaccess.thecvf.com/WACV2025_workshops/GeoCV