👋 About Me

I am a Ph.D. student in Geography at the University of Wisconsin–Madison, affiliated with the Spatial Computing and Data Mining Lab. I completed my M.S. in Geography (Cartography/GIS, thesis) at UW–Madison in May 2026 and am also pursuing an M.S. in Computer Sciences.

My research focuses on geospatial artificial intelligence (GeoAI) and remote sensing for environmental hazards, particularly active-fire detection, wildfire-progression forecasting, and post-fire burned-area and damage assessment. I integrate multi-source satellite observations with weather, terrain, and land-cover data to develop spatiotemporal and physically informed learning methods.

I also work on hyperspectral reconstruction, remote-sensing super-resolution, and hydrological time-series modeling, using transformers, graph neural networks, and multimodal deep learning. My goal is to develop generalizable and interpretable Earth-observation methods that support environmental monitoring and decision-making.

🔬 Current Research Topics

• Wildfire early detection: combining GOES and VIIRS observations with remote-sensing super-resolution and active-fire detection.
• Wildfire-progression forecasting: modeling multi-step fire-front evolution with satellite observations and environmental conditions.
• Post-fire mapping: burned-area and burn-severity assessment using bi-temporal imagery and attention-based models.
• Hyperspectral reconstruction: recovering spectral and spatial detail from sparse multispectral observations.
• Hydrological modeling: learning spatial and temporal dependencies in river networks for streamflow imputation and prediction.

📫 Contact: tsui5@wisc.edu
📄 CV · Google Scholar · ResearchGate

🔥 News

• 2026.09 🎓 Began the M.S. in Computer Sciences at UW–Madison.
• 2026.06 🏆 Received the Best Student Paper Award (1st Place) in the CaGIS Student Paper Competition at the 2026 UCGIS Symposium.
• 2026.05 🎓 Completed my M.S. in Geography with a 4.0/4.0 GPA and began my Ph.D. in Geography at UW–Madison.
• 2026 📝 HydroTGNN published in Environmental Modelling & Software.
• 2026 🎉 SpecSwin3D published in IEEE Transactions on Geoscience and Remote Sensing.
• 2025 🎉 SpecSwin published at the GeoSIM Workshop, ACM SIGSPATIAL 2025.
• 2024 📝 BiAU-Net published in International Journal of Applied Earth Observation and Geoinformation.

📝 Selected Publications

Journal Articles

SpecSwin3D: Generating Hyperspectral Imagery From Multispectral Data via Transformer Networks and Curriculum-Based Cascade Training

Tang Sui, Songxi Yang, and Qunying Huang
IEEE Transactions on Geoscience and Remote Sensing, 64, 1–13, Article 5504113, 2026.
Paper

HydroTGNN: Filling Missing Streamflow Data with Graph Neural Network

Tang Sui, Cheng Wang, Yi Zhuang, Songxi Yang, Jichao Fang, and Qunying Huang
Environmental Modelling & Software, 199, Article 106937, 2026.
Paper

GeoAwareDiffSR: Geospatial-Aware Remote Sensing Super-Resolution with Diffusion Priors and Multi-Modal Constraints for Crop Mapping

Songxi Yang, Tang Sui, and Qunying Huang
GIScience & Remote Sensing, 62(1), Article 2597564, 2025.

Advancing Self-Supervised Learning for Building Change Detection and Damage Assessment: Unified Denoising Autoencoder and Contrastive Learning Framework

Songxi Yang, Bo Peng, Tang Sui, Meiliu Wu, and Qunying Huang
Remote Sensing, 17(15), Article 2717, 2025.
Paper

BiAU-Net: Wildfire Burnt Area Mapping Using Bi-Temporal Sentinel-2 Imagery and U-Net with Attention Mechanism

Tang Sui, Qunying Huang, Mingda Wu, Meiliu Wu, and Zhou Zhang
International Journal of Applied Earth Observation and Geoinformation, 132, Article 104034, 2024.
Paper

A Remote Sensing Spectral Index Guided Bitemporal Residual Attention Network for Wildfire Burn Severity Mapping

Mingda Wu, Tang Sui, Bo Peng, Manzhu Yu, and Qunying Huang
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 17, 17187–17206, 2024.
Paper

Conference and Workshop Papers

SpecSwin: Enhancing Hyperspectral Data Generation through Deep Learning and Multispectral Imagery

Tang Sui, Songxi Yang, and Qunying Huang
8th ACM SIGSPATIAL International Workshop on Geospatial Simulation (GeoSIM), pp. 85–93, 2025.
Paper

Self-supervised Pretraining with Edge Guidance for Building Damage Assessment

Songxi Yang, Bo Peng, Tang Sui, and Qunying Huang
7th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery (GeoAI), 2024.
Workshop Program

📄 CV · Complete publication list on Google Scholar

🎖 Honors & Awards

• Best Student Paper Award (1st Place), CaGIS Student Paper Competition, 2026 UCGIS Symposium, 2026.
• Student Paper Competition Finalist and Travel Award, UCGIS, 2026.
• Finalist, AAG Remote Sensing Specialty Group Student Illustrated Paper Competition, 2026.
• Student Research Grants Competition — Conference Presentation Funds, UW–Madison Graduate School, 2026.
• Summer Geography Graduate Scholarship, UW–Madison Department of Geography, 2026.
• 4th Place, AAG Remote Sensing Specialty Group Student Competition, 2025.
• First Prize, National UAV Photogrammetry Virtual Simulation Competition, 2022.

📖 Education

University of Wisconsin–Madison, Madison, WI, USA

• Ph.D. in Geography, May 2026 – Present.
• M.S. in Computer Sciences, September 2026 – Present.
• M.S. in Geography (Cartography/GIS, thesis), August 2024 – May 2026. GPA: 4.0/4.0.

Tongji University, Shanghai, China

• B.Eng. in Surveying and Mapping, September 2019 – July 2024.

🤝 Academic Leadership & Service

• Primary organizer and chair, GeoAI and Deep Learning Symposium: AI for Earth Observation, AAG Annual Meeting, 2026.
• Co-organizer, The Frontiers of GeoAI, Symposium on Spatiotemporal Data Science, 2026.
• Ad hoc reviewer for Remote Sensing of Environment, Bioresource Technology, and Remote Sensing Applications: Society and Environment.

🧑‍🏫 Teaching

University of Wisconsin–Madison

• Teaching Assistant, GEOG 578: GIS Applications — Fall 2025 and Fall 2026.
• Teaching Assistant, GEOG 573: Advanced Geocomputing — Spring 2026.
• Grader / instructional support, GEOG 170: Our Digital Globe — Spring 2025.

💬 Selected Talks

• GeoAI for Earth Observation and Wildfire Management, Symposium on Spatiotemporal Data Science, 2026.
• Quasi-Hyperspectral Image: Constructing Hyperspectral Imagery from Multispectral Data Using Convolutional Neural Networks, AAG Annual Meeting, Detroit, 2025.
• Empowering Urban Wildfire Burnt Area Detection with Deep Learning, Global Smart Cities Summit & International Conference on Urban Informatics, Hong Kong, 2023.

💻 Research Experience & Internships

Graduate Research Assistant, Spatial Computing and Data Mining Lab, UW–Madison — August 2024 – Present
Develop GeoAI methods for wildfire monitoring, environmental time-series modeling, hyperspectral reconstruction, and remote-sensing super-resolution using satellite, weather, terrain, land-cover, and sensor-network data.

Chinese Academy of Agricultural Sciences (CAAS) — 2021
Collected crop-cycle data, classified Sentinel-2 NDVI curves, and analyzed agricultural observations using Google Earth Engine.

Tongji University, College of Surveying and Geo-informatics — 2021–2022
Conducted GNSS surveying, built 3D GIS applications, and analyzed geospatial data.

Outside of Research

I enjoy playing soccer outside of research. One of my favorite moments: a strike from the center circle on a FULL-SIZE pitch, against professional soccer players (some of whom went on to join China’s youth national teams).

Last updated: October 2026.