Zhu Lab

#AI and #DataScience in #EarthObservation for #SocialGood #RemoteSensing #MachineLearning #Urbanization #SDGs @TU_Muenchen. We strive for #OpenScience!

[Xiaoxiang ZHU @xiaoxiang_zhu]

#EarthNets: Empowering #AI4EO
our #EarthNets Platform developed @Zhu_XLab @TU_Muenchen got accepted @IEEE_GRSS #GRSM.
paper: https://t.co/NGAs7KnfY8
platform: https://t.co/iF9panSZOc

Main features: 500+ datasets with 10+ attributes, 50+ SOTA models, and 10+libraries

[Xiaoxiang ZHU @xiaoxiang_zhu]

🚨 Our new paper introduces #FloodCast, a scalable AI-driven system for large-scale flood modeling using satellite data. 🌊🌍
Paper: https://t.co/EZ6bwhr50J
Code: https://t.co/thWLGzUC3d
#FloodModeling #AI #RemoteSensing #Hydrodynamics #ClimateTech

[Xiaoxiang ZHU @xiaoxiang_zhu]

🚀 Our new paper on #PolyGNN, a graph neural network for 3D building reconstruction from point clouds, is out now! 🌍🏙️ It's scalable, accurate, and ready for real-world applications.
paper: https://t.co/uE9lwdusmy
code: https://t.co/6hWh0Se6JR
#AI #EO #3D #UrbanModeling

[Xiaoxiang ZHU @xiaoxiang_zhu]

Excited to dive into #MachineLearning for #EarthObservation? 🌱🌐 Check out this free MOOC: "Introduction to Machine Learning for Earth Observation" being part of #ML4Earth: https://t.co/6QBxxDZnJi
link to project: https://t.co/O83kqgFCzS

[Zhu Lab @Zhu_XLab]

congrats, Sugandha!

[Xiaoxiang ZHU @xiaoxiang_zhu]

Happy to share that my PhD student Sugandha Doda successfully defended her PhD thesis on "Population Estimation Utilizing Earth Observation Data"! warmest congrats to Dr. Doda for this great achievement! 🌍📊 #PhDDefense #PopulationEstimation #Congratulations #AI4EO

[Zhu Lab @Zhu_XLab]

Congratulations, @koheidler

[Xiaoxiang ZHU @xiaoxiang_zhu]

Huge congratulations to Dr. Heidler @koheidler for his outstanding work on #deeplearning, #remotesensing & #polarregions. Proud of your achievements and looking forward to your future contributions! 🌍❄️📊
#PhD #ClimateChange #RemoteSensing #AI4EO #ProudSupervisor

[OrionLab @OrionLab_NOA]

Join @shanzhao333 Orals #1 at the 2nd #ML4RS Workshop at #ICLR2024 on Causal GNNs for #WildfireDangerPrediction.

🗓️TOMORROW at 12:15 Vienna (Room Lehar 4)!

Not in Vienna? Check out the paper on arXiv: https://t.co/DvkPbyty1A
@Zhu_XLab @Orion_lab_NOA @isp_uv_es

#SeasFire #ESA

[Xiaoxiang ZHU @xiaoxiang_zhu]

🚀 Perspectives on Earth and Climate Foundation Models! 🌍 Check out our new paper, "On the Foundations of Earth and Climate #FoundationModels," where we define 11 essential features to guide research towards the ideal FM.
👉 Read it here: https://t.co/717gegfCjL

[Xiaoxiang ZHU @xiaoxiang_zhu]

🌍Exciting updates from our @BMWK @DLR_en funded Center #ML4Earth ! On April 23, we held our first in-person progress meeting @TU_Muenchen. 🚀
@grosse_guido @MiguelDMahecha @jlbamber @UniBonn @DLR_en @UniLeipzig @AWI_de @ai4eo_de @Zhu_XLab #AI4EO #Climate

https://t.co/XFxWHWFSpp

[Xiaoxiang ZHU @xiaoxiang_zhu]

Big News: we share the insights on global buildings from space today! Explore the unseen with #GlobalOpenBuildingMap! We close the information gap of >50% of buildings across the globe.
#AI4EO #Sustainability #RenewableEnergy #SDGs #urbanization
paper: https://t.co/eFy2rRXyD5

[Xiaoxiang ZHU @xiaoxiang_zhu]

🚀 Dive into the #VisionLanguageModels in #RemoteSensing with our latest paper: "Vision-Language Models in Remote Sensing: Current Progress and Future Trends"
🌐🔍Read the full paper here:
https://t.co/M24ynuhNWc
#AI4EO #Geospatial #NLP #earthobservation

[Xiaoxiang ZHU @xiaoxiang_zhu]

Our team is looking for a senior scientist/manager who is interested in working with us to help shape the profile of our lab in teaching and research:
https://t.co/bty02nE37p

[Zhu Lab @Zhu_XLab]

The PI of our lab @xiaoxiang_zhu will give a keynote @ECMWF @esa #ML4ESOP workshop in May. Stay tuned. @ai4eo_de

[ECMWF @ECMWF]

Keynote presentations are from @xiaoxiang_zhu on Machine Learning for Earth Observation & Beyond, and from Martin Schultz on what will come next in #AI for Weather models, Climate models & Earth system models.
Full programme and to register by 3 May ➡️ https://t.co/I5YpoyJQB0

[Ivica Obadic @ObadicIvica]

Happy to share that our "Contrastive Pretraining for Visual Concept Explanations of Socioeconomic Outcomes" paper (https://t.co/dOFvGZjwSL) was accepted at the #EarthVision workshop at #CVPR2024.
with @Alevering1, Lars Penning, Dario Oliveira @diego_marcos_g, and @xiaoxiang_zhu

[Xiaoxiang ZHU @xiaoxiang_zhu]

How significant would #urbanization impact urban tree coverage? Jianhua extracted #UrbanTreeCoverage of more than 800 cities in North America via #AI4EO, from which prevalent positive indirect effect of urbanization on UTC was revealed.
Link to our paper: https://t.co/MvKLvegeWN

[Xiaoxiang ZHU @xiaoxiang_zhu]

wish for one #EO #FoundationModel for all types of input data with different modalities #SAR #Optical, with different # of bands or even for unseen sensors? Check our new #NeuralPlasticity Inspired model -#DOFA.
paper: https://t.co/SwcqzTFTKD
code: https://t.co/xlMq5IHfNU

[Xiaoxiang ZHU @xiaoxiang_zhu]

#Causality instead of #correlation? We integrate causality with #GraphNeuralNetworks (#GNNs) to better understand the interplay of different factors for #wildfire danger prediction. @Zhu_XLab @ai4eo_de @OrionLab_NOA @isp_uv_es
Paper @TheICLR #ML4RS: https://t.co/uxLcqJZoDk

[Xiaoxiang ZHU @xiaoxiang_zhu]

#PopulationDensity estimation from #Space? In our recent paper @Zhu_XLab led by Sugandha, we use #opendata to do so with #explainable #AI4EO. The approach is #scalable! #AI4Good #SDGs
@ObadicIvica @Wang_SAR @ai4eo_de
paper: https://t.co/0WuQerjPXm
code: https://t.co/yIa2GxAw6F

[Xiaoxiang ZHU @xiaoxiang_zhu]

Curious about #Referring #RemoteSensing Image #Segmentation? It aims at segmenting out the objects from a RS image, given natural language expression. We introduce this new #NLP in #EO task, a dedicated dataset #RefSegRS, and a #LGCE module.
paper: https://t.co/BMPA1fW2Ca

[Xiaoxiang ZHU @xiaoxiang_zhu]

Interested in getting an overview on #building extraction from #RemoteSensing imagery? Qingyu complied a comprehensive survey, recently published @IEEE_GRSS #TGRS - RS offers a perspective to complement incomplete #OSM building data.

Link to paper: https://t.co/GMt5OtXxRq