More about HKUST
Building Marine Foundation Models: Problem Formulation, Models and Applications
PhD Thesis Proposal Defence Title: "Building Marine Foundation Models: Problem Formulation, Models and Applications" by Mr. Ziqiang ZHENG Abstract: The marine ecosystem is the most productive of all ecosystems and shares immense ecological, social, and economic value. Performing marine study plays a significant role in protecting the marine environment and understanding marine science. The marine research involves the study of marine biology, oceanography, and environmental science through the lens of visual data, enabling scientists and researchers to observe, document, and analyze the vast and mysterious creatures beneath the water's surface. Existing marine studies highly depend on describing and analyzing the collected image/video data based on in-situ marine/underwater surveying approaches. There are two main limitations for existing marine studies: 1) they cannot support a very large scale data collection and data scarcity has become one of the important factors that hinder the further development of the marine analysis; 2) further data analysis procedure still requires many human labors, time costs, and is also limited to specific biology users. Recent foundation models have achieved great success, driven by a significant scale of training data and powerful networks. Such foundation model recipe leads to efficient and flexible models, supporting a wide spectrum of downstream visual analysis tasks. However, few attempts have been explored in the marine field and we aim to build effective and efficient marine foundation models. Furthermore, most existing marine visual analysis algorithms are mainly data-driven, specially designed for some tasks and pre-defined conditions. In this work, we try to formulate the basic tasks for marine visual understanding and explore the solutions for large-scale, efficient, repeated surveying, monitoring and further analysis procedures. We first identify the specific and universal challenge of the aquatic environments, the visibility degradation due to the water medium. We propose to do the underwater image/video enhancement to alleviate the visibility issue. We then perform panoptic understanding of the marine world comprehensively, formulating how to do the marine visual analysis in theory and comprehensively. Based on the intrinsic properties, we design different foundation models: CoralSCOP for coral reef segmentation and MarineInst for marine creature identification. We extend our research from image domain to the video field, ensuring 3D scene reconstruction, understanding and 4D animation. We have also reviewed the existing marine datasets and analysis platforms designed for marine analysis. The detailed and hierarchical discussions about potential applications of built marine foundation models are also included. Finally, we discuss the insightful future directions for promoting the marine visual analysis. Date: Monday, 6 January 2025 Time: 10:00am - 12:00noon Venue: Room 3494 Lifts 25/26 Committee Members: Prof. Sai-Kit Yeung (Supervisor) Prof. Hongbo Fu (EMIA) (Chairperson) Dr. Qifeng Chen Dr. Dan Xu