4 papers
Mobility-Embedded POIs: Learning What A Place Is and How It Is Used from Human Movement
Maria Despoina Siampou, Shushman Choudhury, Shang-Ling Hsu +2
Recent progress in geospatial foundation models highlights the importance of learning general-purpose representations for real-world locations, particularly points-of-interest (POI…
TRAJGANR: Trajectory-Centric Urban Multimodal Learning via Geospatially Aligned Neural Representations
Maria Despoina Siampou, Gengchen Mai, Ni Lao +4
Multimodal self-supervised learning (MSSL) has emerged as a key paradigm for pretraining geospatial foundation models. However, existing geospatial MSSL methods are mainly designed…
S2Vec: Self-Supervised Geospatial Embeddings for the Built Environment
Shushman Choudhury, Elad Aharoni, Chandrakumari Suvarna +4
Scalable general-purpose representations of the built environment are crucial for geospatial artificial intelligence applications. This paper introduces S2Vec, a novel self-supervi…
Scalable Learning of Segment-Level Traffic Congestion Functions
Shushman Choudhury, Abdul Rahman Kreidieh, Iveel Tsogsuren +3
We propose and study a data-driven framework for identifying traffic congestion functions (numerical relationships between observations of traffic variables) at global scale and se…