Turning Earth Observations to Natural Language Dialogues
This talk covers the integration of recent GenAI technologies (Large Language Models) for earth observation and climate data. We will cover models developed by MBZUAI and its collaborators in recent years for improved analysis of GeoSpatial data using natural language conversations. These models can process multisensory inputs i.e., RGB, SAR, Mult-spectral and Temporal sequences and allow additional capabilities such as visual grounding and change detection. The talk will also cover applications of these models to specific use cases, e.g., damage assessment due to extreme weather events and tracking urban heat-island phenomena.
Dr. Salman Khan
Associate Professor at Mohamed Bin Zayed University of Artificial Intelligence
Dr. Salman Khan is an Associate Professor at Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI) and an honorary faculty member at the Australian National University (ANU). He works on multimodal learning algorithms for various applications, with a specific focus on earth observation and climate science. He has been a recipient of several prestigious scholarships, including Fulbright and IPRS, and served as a program committee member for several premier conferences where he has won multiple outstanding reviewer awards. He was the guest editor for IEEE TPAMI, MDPI Remote Sensing and acts as area chair for premier AI conferences such as NeurIPS, CVPR, ICML and ICCV. He, alongside his collaborators and students, won the best paper award in ICPRAM 2020, best paper honorable mention at ACCV 2024 and top ranks in CVPR-NTIRE 2019 and 2021 challenges on image enhancement. He has published over 100+ technical papers in top AI journals and conferences.
Related Events
