Cyber-GIS Center for Advanced Digital and Spatial Studies

🏆 Spatial AI Challenge 2025–26: First Place Project 🏆

We are excited to highlight the first-place winner of the Spatial Challenge 2025–26, hosted on the I-GUIDE Platform.

🌆 MURD-ViT: Urban Retrofitting Detection with Vision Transformer
By Raj Bhattarai and Fangzheng Lyu (Virginia Tech)

This project introduces an AI-driven framework for detecting and analyzing urban retrofitting using street-view imagery, socioeconomic data, and Vision Transformer models. By identifying where, when, and what types of retrofitting occur, the project provides scalable tools for evaluating climate impacts, including urban heat island mitigation, while also examining equity in access to sustainable urban improvements.

Hosted on the I-GUIDE Platform, the Spatial AI Challenge brings together researchers and innovators from around the world to develop responsible, reproducible, and impactful AI solutions guided by FAIR and open principles.

🔗 Learn more about the challenge and winning projects: https://i-guide.io/spatial-ai-challenge-2025-26/

🌟 New publication from the NSF I-GUIDE Team!

This publication is from Computers, Environment and Urban Systems, "Customized decision-making support per location's characteristics of spatial accessibility: A case study of emergency healthcare in New York City" by Jinwoo Park, Fangzheng Lyu, Alexander Michels, Ian Zhang, and Shaowen Wang.

📃 Read the full paper here 👉 https://doi.org/10.1016/j.compenvurbsys.2026.102458

🧠 TODAY AT 11 AM CT!

Join Qiusheng Wu of the University of Tennessee for an upcoming VCO on building accessible, cloud-based GeoAI education. This session will explore open-source teaching modules that guide learners through end-to-end workflows—from data access to deep learning—using scalable, reproducible tools on the I-GUIDE Platform.

🔗 Link: https://i-guide.io/i-guide-vco/open-source-geoai-education-reproducible-workflows-for-geospatial-ai/

🌟 I-GUIDE Summer School 2026 Project Spotlight 🌟

This summer, participants in the I-GUIDE Summer School will tackle real-world challenges at the intersection of , remote sensing, and open .

🛰️ Detecting and Correcting Spatial Bias in VGI Using Remote Sensing

Led by Fangzheng Lyu of Virginia Tech University, this project investigates how multimodal remote sensing data—including satellite imagery and LiDAR—can be used to assess and improve the quality of Volunteered Geographic Information (VGI) datasets such as OpenStreetMap. The team will explore geographic patterns of data completeness and accuracy, develop methods to detect discrepancies in mapped features, and design scalable AI workflows to enhance VGI quality in data-sparse regions.

Through interdisciplinary collaboration, participants will contribute to building more reliable geospatial data resources that support urban analytics, disaster response, environmental , and other critical applications.

🔗 Learn more about the Summer School projects: https://i-guide.io/summer-school/summer-school-2026/summer-school-2026-projects/

🌟 I-GUIDE Platform Knowledge of the Week!

For the week of June 1, "Notebooks: the Qiusheng Wu collection for GeoAI Education" by Qiusheng Wu is the knowledge element of the week.

Qiusheng Wu, from the University of Tennessee and an I-GUIDE UCGIS Community Champion, has developed a series of Notebooks that are now available on the I-GUIDE Platform. Explore the collection and learn about topics such as Change Detection, Instance Segmentation, or Vision Language Models. On Wednesday June 3, this collection will be the focus of a VCO offered by Dr. Wu, Open-Source GeoAI Education: Reproducible Workflows for Geospatial AI. Join us!

🔗 Access the element here: https://platform.i-guide.io/search?keyword=qiusheng&type=any

🔗 Sign up for the VCO: https://i-guide.io/i-guide-vco/open-source-geoai-education-reproducible-workflows-for-geospatial-ai/

🌐 Missed our recent I-GUIDE VCO? The recording is now available!

Thank you to Jungha Woo of Purdue University, Erick Li of the University of Illinois Urbana-Champaign, and Michael Englert of Utah State University for an insightful session on Scaling Geospatial Insights: A Reproducible API-Driven Framework for Dam-Failure Risk Assessment.

The presentation showcased a scalable framework that transforms expert-led GIS workflows into a reproducible, PostGIS-backed REST API and dashboard system. The session demonstrated how hazard, infrastructure, environmental, and social vulnerability datasets can be integrated into an automated geospatial analysis pipeline, supporting transparent and consistent dam-failure risk assessment across large geographic regions.

🎥 Watch the recording here: https://youtu.be/d3qt9ghhzu8?si=EIQySxjR1P9OCtHa

Scaling Geospatial Insights: A Reproducible API-Driven Framework for Dam-Failure Risk Assessment Expert-led geospatial analysis is the foundation of dam-failure con...

🌐 Missed our recent I-GUIDE VCO? The recording is now available!

Thank you to Krasen Samardzhiev of Lampata Ltd for an insightful session on Open Science and Reproducible Workflows with Urban Taxonomy and OSPD.

The presentation explored challenges and opportunities in FAIR and Open Science through the Urban Taxonomy project and the Open Science Persistent Demonstrator (OSPD). The session highlighted approaches for accessing, visualizing, and reproducing scientific data and code, while showcasing tools and workflows that support transparent, scalable, and collaborative geospatial research.

🎥 Watch the recording here: https://youtu.be/WlVbx2BBI9c si=0yWR8FpUkiTrVJX4

Open Science and Reproducible Workflows with Urban Taxonomy and OSPD Despite significant advancements in FAIR and Open Science via new platforms (EarthCODE), tools (Jupyter, Quarto), and even journals (Urban Planning B: Data a...

📅 Upcoming I-GUIDE Virtual Consulting Offices (VCOs)

Join us this June and July for a series of engaging VCO sessions exploring the latest in data science, AI, and open science :

🛰️ Open-Source GeoAI Education: Reproducible Workflows for Geospatial AI
Discover how Qiusheng Wu of the University of Tennessee, Knoxville is developing open-source, cloud-enabled teaching modules for GeoAI, using tools like SamGeo, deep learning, and Jupyter Notebooks to support scalable and reproducible geospatial AI education.

🌳 Open GeoAI for Urban Tree Canopy Mapping in U.S. Cities
Learn how Yi Qi of the University of Southern California is developing an open and reproducible GeoAI framework for urban tree canopy mapping using USDA NAIP imagery and open-source image segmentation workflows. This session highlights scalable approaches for remote sensing education, model development, and urban forest management across cities.

🏙️ RegionLM: Integrating Point, Line, and Polygon Context for Urban Representation Learning
Explore how Yao-Yi Chiang, Yijun Lin, and Zekun Li are advancing urban representation learning through RegionLM, a geospatial AI framework that generates contextual region embeddings from OpenStreetMap features. This session highlights how semantic, spatial, and topological information can be combined to support applications such as mobility modeling, neighborhood analysis, and trajectory anomaly detection.

🔗 Learn more & register: https://i-guide.io/i-guide-vco/

🌊 Today at 11 AM CT!

Join Jungha Woo of Purdue University, Erick Li of the University of Illinois Urbana-Champaign, and Michael Englert of Utah State University for an upcoming VCO on scalable and reproducible geospatial workflows for dam-failure risk assessment.

🔗 Link: https://i-guide.io/i-guide-vco/scaling-geospatial-insights-a-reproducible-api-driven-framework-for-dam-failure-risk-assessment/

🌟 I-GUIDE Platform Knowledge of the Week!

For the week of May 25, "Vulnerabilities Downstream of Dams" by Michael Englert and Nicolas Holden is the knowledge element of the week.

To understand the characteristics of populations living downstream from aging dams, three nation-wide datasets in SETS categories (social, ecological, and technological). Focusing initially on the state of Utah, the datasets enable comparison of state-regulated high-risk dams and were combined in R to produce a vulnerability matrix, replicated on the I-GUIDE platform as an interactive dashboard for querying vulnerability information.

🔗 Access the element here: https://platform.i-guide.io/datasets/a660bd31-5cdd-4f24-8a92-5d0366ba8b13