Keynote Speakers

Abstract: TBA

Lorenzo Torresani is a Professor and a President Joseph E. Aoun Chair at Northeastern University, Khoury College of Computer Sciences. He is the author of the first methods for 3D reconstruction of non-rigid objects and influential architectures for image and video analysis (e.g., Classemes, C3D, TimeSformer). His current research interests are in video-language models. He received his Ph.D. in Computer Science from Stanford University in 2005. In the past, he held a tenured Professor position at Dartmouth College and was a Research Director at Meta, Fundamental AI Research (FAIR). He also worked at Microsoft Research, Like.com, and Digital Persona. He is the recipient of multiple awards, including a CVPR Best Student Paper prize, a National Science Foundation CAREER Award, a Google Faculty Research Award, three Facebook Faculty Awards, and a Fulbright U.S. Scholar Award.

Abstract: TBA

Georgia Gkioxari is an assistant professor of Computing + Mathematical Sciences at Caltech and a Hurt Scholar. She also spends time with the FAIR Perception team at Meta AI. From 2016 to 2022, she was a research scientist at Meta's FAIR team. Georgia Gkioxari received her PhD from UC Berkeley, where she was advised by Jitendra Malik. She did her bachelors in ECE at NTUA in Athens, Greece, where I worked with Petros Maragos.

In this talk, I will explore the pivotal role of AI in tackling the challenges of visual understanding in the marine environment. In particular, I will introduce the concept of marine AI foundation models—large-scale pre-trained systems designed for marine visual understanding. These models aim to capture the unique characteristics of underwater environments and serve as a universal system for various visual understanding tasks, such as coral reef monitoring, biodiversity assessment, marine species monitoring, and underwater navigation. Additionally, I will discuss several ongoing projects focused on achieving accurate underwater 3D mapping. The research I will present contributes to interdisciplinary efforts across maritime planning, large-scale seafloor surveying and cleaning, fishery management, marine life monitoring, underwater trash cleaning, and advancing AI applications in marine science and technology.

Sai-Kit Yeung is a Professor at the Division of Integrative Systems and Design (ISD), the Department of Computer Science and Engineering (CSE) and the Department of Ocean Science (OCES) at the Hong Kong University of Science and Technology (HKUST). Before joining HKUST, he was an Assistant Professor at the Singapore University of Technology and Design (SUTD) and founded the Vision, Graphics and Computational Design (VGD) Group. During his time at SUTD, he was also a Visiting Assistant Professor at the Computer Science Department at Stanford University and the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT. Prior to that, he had been a Postdoctoral Scholar in the Department of Mathematics, University of California, Los Angeles (UCLA). Prof. Yeung’s research interests include 3D vision and graphics, content generation, fabrication, novel computational techniques and integrative systems for marine-related problems. He has published extensively in premier computer vision and graphics venues, including numerous full oral papers in CVPR, ICCV, and AAAI. His work has received best paper honorable mention awards at ICCP 2015 and 3DV 2016. Prof. Yeung has been actively serving as a senior committee member in major AI, computer vision, computer graphics, and robotics conferences. These include roles as a Senior Program Committee member for IJCAI 2021 and AAAI 2021 and 2022, an Area Chair for ICCV 2023, CVPR 2023, 2024, and 2025, NeurIPS 2024, and ICML 2025, as well as a Lead Area Chair for CVPR 2026. He also serves as an Associate Editor (Robot Learning) for ICRA 2024, 2025, and 2026. He was Course Chair for SIGGRAPH Asia 2019 and regularly contributes as a Technical Papers Committee member for Eurographics, SIGGRAPH, and SIGGRAPH Asia. He is currently an Associate Editor of the ACM Transactions on Graphics (TOG).

Medical AI research shows great promise for changing healthcare. AI can help improve diagnoses, personalize treatments, and speed up drug discovery. However, putting this research into daily hospital practice is still difficult. In this talk, I will discuss challenges to application of AI to medicine. We will begin by exploring the potential of AI to revolutionize medicine, and discuss the general challenge to applying it to the frontline in general. Then, I will discuss challenges that I see in context of Thailand, where hospital infrastructure, rules, and resources can limit AI use. Nonetheless, some small successes of AI applications, like those at Siriraj Hospital, offer some hope. I will conclude by offering some thoughts on what we can do to stack the odds in our favor, for using AI to improve lives for everyone.

Dr. Ronnachai Jaroensri is a lecturer at the Faculty of Medicine, Siriraj Hospital, Thailand. Until mid-2025, he was a researcher in Google Health working on advancing medical multimodal capabilities of Gemini and Gemma, Google's generative AI models. He focused on developing the vision-language ability of LLMs to create a performant system that can interact with users in a more natural way. Before Google, he completed his PhD study under Professor Freédo Durand at MIT CSAIL, working on using synthetic data to solve various computer vision and signal processing problems. Previously, he attended Stanford University for a bachelor's degree in Physics.