Self-supervised and multimodal embeddings that align vision, language, and audio for robust learning across sensors and domains.
Explore →Multimodal models for species, habitats, and environmental change—from natural-world imagery to conservation-ready monitoring.
Explore →Planet-scale geospatial ML: unified embeddings, cross-view localization, and structured reasoning over satellite, aerial, and ground imagery.
Explore →Generative models for images, 3D scenes, and multimodal earth data across scales and sensors.
Explore →We are a computer vision lab at Washington University in St. Louis, led by Nathan Jacobs. We develop representation learning, geospatial AI, and generative modeling methods for satellite, aerial, and ground imagery, with extensions to audio, text, and structured data. We validate these methods on real-world challenges in biodiversity and conservation, agriculture, environmental monitoring, the built environment, and transportation safety.