Speaker Bio

Dr. Leifman is a Research Scientist at Google Research, working on Google Earth AI, where he develops vision-language foundation models for satellite and aerial imagery. His previous industry experience includes building computer vision systems for Amazon Go and Amazon One at Amazon Research, and developing ML/CV projects in the healthcare domain at Google Health and Verily. Dr. Leifman holds a Ph.D. in Electrical Engineering from the Technion and completed a Postdoctoral Fellowship at the MIT Media Lab.

Homepage: https://research.google/people/george-leifman-2

Title: EarthAI and Remote Sensing Foundation Models at Google
Abstract: This talk will present EarthAI, a Gemini-powered Geospatial Reasoning Agent designed to address complex, fragmented data about our planet. It combines foundational models for Imagery (Remote Sensing), Population Dynamics, and Environment (Weather, Floods, Cyclones) to provide non-experts with clear, actionable insights in minutes. The agent acts as an intelligent orchestrator, enabling multi-step, rigorous spatial analysis through natural language, as demonstrated by the “Ask Google Earth” experimental capability.
The second part of the talk will focus on remote sensing foundation models and their application in building robust, generalizable models for geospatial tasks, such as visual catalog browsing and retrieval, change detection, open vocabulary object detection, and counting.