projects

VidTAG: Temporally Aligned Video to GPS Geolocalization with Denoising Sequence Prediction at a Global Scale

Determining the GPS location of every frame of a ground video, by direct matching
(with National Geospatial Intelligence Agency (NGA))

CityGuessr: Video Geolocalization through Classification

Identifying the city a video is taken in
(with National Geospatial Intelligence Agency (NGA))

MedRoute: RL-Based Dynamic Specialist Routing in Multi-Agent Medical Diagnosis

Flexible Multi-Agent System for Medical Diagnosis

GeoDecoder

Query Based Worldwide Image Geo-Localization Using Hierarchies and Scenes
(with Army Research Lab (ARL))

GAEA: Conversational VLM Assistant for worldwide image geolocalization

(with Lockheed Martin)

Geo-HiPO

Worldwide Image Geolocalization via Hierarchical Preference Optimization
(with NSF)

Self-supervised Same-view Video Geolocalization

(with NSF)

Seismic Waveform Reconstruction

Reconstructing gaps in seismic waveforms
(with AFTAC, US Air Force)

NeuroChat

VLM for Tumor Grounding

Eye in the Sky

Exoplanet Detection