Safwan Wshah
Associate Professor, Department of Computer Science
Graduate Program Director, Department of Computer Science
Director, Vermont Artificial Intelligence Laboratory (VaiL)
Core Member, Vermont Complex Systems Institute
Affiliate, Gund Institute for Environment
Innovation E415
82 University Place
Burlington, VT 05405
United States
- Ph.D. in Computer Science and Engineering, University at Buffalo, State University of New York, USA, 2012.
- M.Sc. in Communication Engineering, University of Jordan, Amman, Jordan, 2005.
- B.Sc. in Electronics Engineering, Princess Sumaya University for Technology, Amman, Jordan, 2001.
Department of Computer Science
Gund Institute for Environment
Vermont Complex Systems Institute
Areas of expertise
Artificial Intelligence and Machine Learning, Deep Learning and Computer Vision, Geospatial AI and Remote Sensing, AI for Medicine and Biomedical Discovery, Cross-View Geo-Localization and Multi-View Learning, Multimodal Learning and Vision-Language Models, Generative AI and Foundation Models, Robust and Privacy-Preserving Artificial Intelligence, AI for Transportation Safety and Environmental Monitoring
BIO
Dr. Safwan Wshah is an Associate Professor of Computer Science at the University of Vermont and Director of the Vermont Artificial Intelligence Laboratory (VaiL). His research spans artificial intelligence, machine learning, computer vision, generative AI, and multimodal learning, with an emphasis on advancing foundational AI methodologies and developing intelligent systems to address complex real-world challenges. His work bridges fundamental machine learning research with interdisciplinary applications in geospatial intelligence, medicine, biomedical discovery, transportation, and environmental science.
A central theme of Dr. Wshah's research is localization and spatial intelligence across diverse environments, sensing modalities, and applications. His research encompasses wireless sensor networks, indoor and outdoor localization, visual geo-localization, and medical device positioning. He develops advanced machine learning methods that enable intelligent systems to understand spatial relationships, integrate information from multiple sensors and perspectives, and accurately estimate the locations of objects and devices in complex environments. His work emphasizes robustness, generalization, and privacy-preserving learning to support reliable AI systems under diverse operating conditions.
In computer vision and geospatial AI, Dr. Wshah focuses on cross-view geo-localization, multi-view visual understanding, object-level localization, and spatial reasoning using ground-level, aerial, and satellite imagery. His laboratory develops novel deep learning methods, foundation models, and generative AI approaches to overcome the significant visual and geometric differences between viewpoints. His recent research explores vision-language models, diffusion-based image synthesis, and multimodal spatial reasoning to advance the ability of AI systems to understand, interpret, and reason about complex geographic environments.
Dr. Wshah is also actively engaged in AI for medicine and biomedical discovery, contributing to interdisciplinary research that applies advanced generative AI, foundation models, and agentic AI to therapeutic peptide and protein design, including the discovery and optimization of antimicrobial peptides. His laboratory also develops intelligent medical imaging and robotic positioning methods, particularly AI-driven C-arm positioning and automation for image-guided medical interventions. These efforts combine computer vision, vision-language models, and spatial reasoning to improve the precision, efficiency, and safety of medical procedures, with the goal of reducing unnecessary imaging, procedure time, and radiation exposure.
His research also extends to AI applications in transportation safety, environmental monitoring, and autonomous systems. His group develops computer vision, remote sensing, and multimodal AI solutions for roadway infrastructure assessment, transportation safety, and environmental monitoring, including the detection and analysis of harmful algal blooms. These efforts demonstrate how advances in foundational AI and machine learning can be translated into practical solutions addressing important scientific, engineering, and societal challenges.
Dr. Wshah leads and contributes to an active, federally funded research program supported by major agencies, including the National Science Foundation (NSF), National Institutes of Health (NIH), and National Oceanic and Atmospheric Administration (NOAA). He collaborates extensively with researchers at universities, medical centers, national research institutions, and other organizations across the United States. These collaborations bring together expertise in computer science, engineering, medicine, biomedical sciences, and environmental science, spanning both foundational machine learning research and interdisciplinary AI applications. Through these partnerships, he advances novel AI algorithms and methodologies while translating research innovations into practical solutions with broad scientific and societal impact.
Dr. Wshah received his Ph.D. in Computer Science and Engineering from the University at Buffalo in 2012. Before joining the University of Vermont, he worked as a research scientist at Xerox Research and Palo Alto Research Center (PARC). He is a recipient of the Fulbright Global Scholar and Fulbright Specialist awards.
Courses
- Machine Learning
- Advanced Machine Learning
- Deep Learning
- Foundations of Large Language Models (LLMs)
Publications
Publications on Google Scholar
Awards and Achievements
- Fulbright Specialist Award (2025).
- Fulbright Global Scholar Award (2023–2024) — U.S. Department of State, international research and academic collaboration.
- Excellence in Research Award (2023) — College of Engineering and Mathematical Sciences (CEMS), University of Vermont.
- PARC Special Award for Contributions to the Most Innovative Project (2015) — Palo Alto Research Center (PARC), Xerox.
Bio
Dr. Safwan Wshah is an Associate Professor of Computer Science at the University of Vermont and Director of the Vermont Artificial Intelligence Laboratory (VaiL). His research spans artificial intelligence, machine learning, computer vision, generative AI, and multimodal learning, with an emphasis on advancing foundational AI methodologies and developing intelligent systems to address complex real-world challenges. His work bridges fundamental machine learning research with interdisciplinary applications in geospatial intelligence, medicine, biomedical discovery, transportation, and environmental science.
A central theme of Dr. Wshah's research is localization and spatial intelligence across diverse environments, sensing modalities, and applications. His research encompasses wireless sensor networks, indoor and outdoor localization, visual geo-localization, and medical device positioning. He develops advanced machine learning methods that enable intelligent systems to understand spatial relationships, integrate information from multiple sensors and perspectives, and accurately estimate the locations of objects and devices in complex environments. His work emphasizes robustness, generalization, and privacy-preserving learning to support reliable AI systems under diverse operating conditions.
In computer vision and geospatial AI, Dr. Wshah focuses on cross-view geo-localization, multi-view visual understanding, object-level localization, and spatial reasoning using ground-level, aerial, and satellite imagery. His laboratory develops novel deep learning methods, foundation models, and generative AI approaches to overcome the significant visual and geometric differences between viewpoints. His recent research explores vision-language models, diffusion-based image synthesis, and multimodal spatial reasoning to advance the ability of AI systems to understand, interpret, and reason about complex geographic environments.
Dr. Wshah is also actively engaged in AI for medicine and biomedical discovery, contributing to interdisciplinary research that applies advanced generative AI, foundation models, and agentic AI to therapeutic peptide and protein design, including the discovery and optimization of antimicrobial peptides. His laboratory also develops intelligent medical imaging and robotic positioning methods, particularly AI-driven C-arm positioning and automation for image-guided medical interventions. These efforts combine computer vision, vision-language models, and spatial reasoning to improve the precision, efficiency, and safety of medical procedures, with the goal of reducing unnecessary imaging, procedure time, and radiation exposure.
His research also extends to AI applications in transportation safety, environmental monitoring, and autonomous systems. His group develops computer vision, remote sensing, and multimodal AI solutions for roadway infrastructure assessment, transportation safety, and environmental monitoring, including the detection and analysis of harmful algal blooms. These efforts demonstrate how advances in foundational AI and machine learning can be translated into practical solutions addressing important scientific, engineering, and societal challenges.
Dr. Wshah leads and contributes to an active, federally funded research program supported by major agencies, including the National Science Foundation (NSF), National Institutes of Health (NIH), and National Oceanic and Atmospheric Administration (NOAA). He collaborates extensively with researchers at universities, medical centers, national research institutions, and other organizations across the United States. These collaborations bring together expertise in computer science, engineering, medicine, biomedical sciences, and environmental science, spanning both foundational machine learning research and interdisciplinary AI applications. Through these partnerships, he advances novel AI algorithms and methodologies while translating research innovations into practical solutions with broad scientific and societal impact.
Dr. Wshah received his Ph.D. in Computer Science and Engineering from the University at Buffalo in 2012. Before joining the University of Vermont, he worked as a research scientist at Xerox Research and Palo Alto Research Center (PARC). He is a recipient of the Fulbright Global Scholar and Fulbright Specialist awards.
Courses
- Machine Learning
- Advanced Machine Learning
- Deep Learning
- Foundations of Large Language Models (LLMs)
Publications
Awards and Achievements
- Fulbright Specialist Award (2025).
- Fulbright Global Scholar Award (2023–2024) — U.S. Department of State, international research and academic collaboration.
- Excellence in Research Award (2023) — College of Engineering and Mathematical Sciences (CEMS), University of Vermont.
- PARC Special Award for Contributions to the Most Innovative Project (2015) — Palo Alto Research Center (PARC), Xerox.
Publications
Selected Publications
- Split and Bridge: Multimodal Generation via Diffusion Bridging. Ahmad Arrabi, X. Zhang, Xingyu Li, and Safwan Wshah. Advances in Neural Information Processing Systems (NeurIPS), 2026.
- Geo²: Geometry-Guided Cross-view Geo-Localization and Image Synthesis. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026.
- GeoFlow: Real-Time Fine-Grained Cross-View Geolocalization via Iterative Flow Prediction. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026.
- Autonomous Skeletal Landmark Localization Toward Agentic C-arm Control. International Journal of Computer Assisted Radiology and Surgery, 2026.
- Agentic Discovery of Non-Canonical Antimicrobial Peptides with AMPGAN v3. ICML Workshop on Generative and Agentic AI for Biology, 2026.
- Spatially-Adaptive Conformal Graph Transformer for Indoor Localization in Wi-Fi Driven Networks. IEEE International Conference on Communications (ICC), 2026.
- UBiGTLoc: A Unified BiLSTM–Graph Transformer Localization Framework for IoT Sensor Networks. IEEE Internet of Things Journal, 2025.
- Cross-View Meets Diffusion: Aerial Image Synthesis with Geometry and Text Guidance. IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2025.
- GeoDTR+: Toward Generic Cross-View Geolocalization via Geometric Disentanglement. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2024.
- Cross-View Geo-Localization via Learning Disentangled Geometric Layout Correspondence. AAAI Conference on Artificial Intelligence, 2023.
- AMPGAN v2: Machine Learning-Guided Design of Antimicrobial Peptides. Journal of Chemical Information and Modeling, 2021.