Quite many from @digigeolab were at the #LBS2025 conference last week presenting our work on #Mobility #GIScience #Geography and #GeoAI, and we wrote a small blog post on it.
Quite many from @digigeolab were at the #LBS2025 conference last week presenting our work on #Mobility #GIScience #Geography and #GeoAI, and we wrote a small blog post on it.
Australian geospatial thought leadership has a massive boner for geoAI right now - as in the JMSU, 1m resolution magicked from 10m data variety. Not the mislabeled statistical model variety.
...and a sane "people, have you not watched this approach fuck up every single thing so far?" approach is just not tolerable.
Better go make a picture of myself as an action figure. Maybe thats how to get hired?
I am looking forward to Isaac Oluoch‘s PhD defence today. Feel free to join us online. It has been a pleasure to serve as his supervisor for his #PSTS MSc thesis and as his Daily Supervisor (w/ Monika Kuffer). #PhilTech #STS #RemoteSensing #GeoAI https://www.itc.nl/events/2025/4/255007/phd-defence-isaac-oluoch-frame-pro-framework-for-the-responsible-and-accountable-mapping-of-deprived-urban-areas
#torchGeo Release v0.7.0
https://www.osgeo.org/community-news/torchgeo-release-v0-7-0/
torchGeo is a #PyTorch domain library designed to make it simple for #machineLearning experts to work with #EO data, providing datasets, samplers, transforms, and pre-trained models specific to #geospatial data.
The quest for a fair TimeGPT benchmark
At the end of yesterday's #TimeGPT for mobility post, we concluded that TimeGPT's trainingset probably included a copy of the popular BikeNYC timeseries dataset and that, therefore, we were not looking at a fair comparison ...
http://anitagraser.com/2025/03/29/the-quest-for-a-fair-timegpt-benchmark/
TIL about a new European data-driven weather forecasting model, the Artificial Intelligence/Integrated Forecasting System (AIFS) that outperforms leading physics-based global numerical weather prediction
https://www.ecmwf.int/en/newsletter/178/news/aifs-new-ecmwf-forecasting-system
GeoAI: State-of-play in 2025: The Innovation Booster #ArtificialIntelligence has released a concise review of #GeoAI in Switzerland. The report highlights significant players, innovations, and developments in the field, showcasing how #AI is being integrated into #GIS...
https://spatialists.ch/posts/2025/03-25-geoai-state-of-play-in-2025/ #GIS #GISchat #geospatial #SwissGIS
Now starting the #tutorial about #trainingDML-AI , by Daniel Moreno #carbonspace #ai #geoai #codesprint25 #standards #mentorstream
Paper Alert
Can we use AI and social media posts to predict pandemic outbreaks?
A study by David Hanny et al. presents a new GeoAI framework that integrates mobile phone data and geo-referenced social media posts to assess their relationship with COVID-19 incidence rates in Rio de Janeiro.
The findings show that social media activity related to COVID-19, tourism and leisure activities strongly correlates with infection rates.
AI tools and open geodata can help humanitarian practitioners in taking informed decisions, optimizing operations, and fostering community resilience with disaster risk reduction and anticipatory action.
We are looking forward to discussing emerging technologies at the Humanitarian Networks and Partnership Weeks 2025:
Session "GeoAI in Action: Practical Tools for Humanitarian Impact"
Sukanya Randhawa
24 Mar 25, 11:00-12:30
Geneve
& online
https://heigit.org/events/geoai-in-action-practical-tools-for-humanitarian-impact-at-hnpw/
We have released #SketchMapTool v2.1!
This version uses a larger training dataset and an updated machine learning model for sketch detection to deliver a higher accuracy in digitizing hand-sketched markings.
Sketch Map Tool is an open-source web app for participatory mapping that enables the creation, digitization, and analysis of paper maps.
Read about SMT v2.1 and access the training dataset & ML model: https://heigit.org/digitizing-maps-with-sketchmaptool-v2-1-enhanced-sketch-detection-and-global-impact-2/
Try it out: https://sketch-map-tool.heigit.org/
This new paper by Steffen Knoblauch et al. explores how students are exposed to crime while commuting to schools in Rio de Janeiro .
Relying on GeoAI techniques, the study reveals inequalities in geographic access to education -- dispute areas increase travel time to the nearest school by 49%.
A concise overview of the study: https://heigit.org/new-paper-crime-associated-inequality-in-geographical-access-to-education-insights-from-the-municipality-of-rio-de-janeiro/
Full paper: https://www.sciencedirect.com/science/article/pii/S0264275125001180?via%3Dihub
Dans le Michigan, des vélos équipés d'un système "léger" de caméra + IA pour cartographier les endroits à risque d'accident et programmer ou adapter des pistes cyclables. L'article aborde les problèmes associés à cette solution HighTech et les réticences observées.
Note : le gouvernement Trump vient d'effacer les pages Web de l'organisme fédéral qui finançait cette initiative.
#GeoAI #Bike #vélo #GéoIA
https://urbanists.social/@TrufiAssoc/113962773447029349
Segment Anything Model Can Not Segment Anything - Assessing AI Foundation Model’s Generalizability In Permafrost Mapping
--
https://doi.org/10.3390/rs16050797 <-- shared paper
--
#GIS #spatial #mapping #remotesensing #foundationmodel #AI #artificialintelligence #zeroshot #segmentation #GeoAI #spatialanalysis #LargeLanguageModel #LLM #SAM #performance #metrics #permafrost #visionmodel #icewedge #Arctic #warming #climatechange #thawslumps #landform #terrainmapping #EuroCrops #agriculture
New video: explore deep learning models in #QGIS with the Deepness plugin. Segment or detect features from #RemoteSensing images. #GeoAI #MachineLearning #GIS youtu.be/xCKVqjeihbU
Deep Learning in QGIS with the...
New video: explore deep learning models in #QGIS with the Deepness plugin. Segment or detect features from #RemoteSensing images. #GeoAI #MachineLearning #GIS https://youtu.be/xCKVqjeihbU
I really enjoyed the interdisciplinary collaboration sparked by the gAia project on predicting #landslides, resulting in our recent #preprint:
"Visualizing uncertainties in landslide susceptibility modeling using bivariate mapping"
la mia video intervista per il festival di "Informatici Senza Frontiere" sul tema mappe ed intelligenza artficiale
#geoai