Achievements
Community
Corporate Projects
Exchange
Field Trips & Visits
Internship & Career
Undergraduate
Office
Research
Seminars & forums
Student Activities
T&M-DDP
Postgraduate
EVMT
Innovation
Entrepreneurship
Sustainability
Engineering
Environment
Air Quality
GBA
PublicPolicy
ENVR
PPOL
Teaching&Learning
Technology
Research and Technology
Greater Bay Area
IIM
Fintech
Research and Innovation
ISD Research Project “Coral Reef Segmentation” Shines at Dubai Future Solutions - Prototypes for Humanity: 2024 Exhibition as HKUST’s First Project Selected
03/12/2024
Thumbnail

Led by Prof. Sai-Kit YEUNG from HKUST’s Division of Integrative Systems and Design (ISD), the Department of Computer Science and Engineering (CSE) and his CSE PhD student Mr. ZHENG Ziqiang, the advanced model for automatic coral reef segmentation has been selected to showcase at Dubai Future Solutions – Prototypes for Humanity, an international program enabling academic innovations that produce real-world impact for humanity, on Nov 18-21, 2024. 
 
Out of 2,700 global applications submitted for the 2024 program, “Coral Reef Segmentation” has been selected as one of the top 100 research-backed projects along with other leading universities globally including ETH Zurich, Imperial College London, Massachusetts Institute of Technology (MIT), University of Cambridge, University of Oxford, and more.

 

This marks the first HKUST project featured in the exhibition, demonstrating the highest potential to solve global issues, with impact at its core, backed by rigorous academic research. Prof. Yeung said, “Our research team is at the forefront of AI technology in coral visual research, positioning us as leaders in coral image analysis worldwide.”

 

Coral reefs, vital for marine biodiversity, are difficult to monitor due to their complex structures and the manual effort required for analysis. This project addresses the challenge by offering an advanced foundation model for the automatic dense segmentation of coral reefs. It uses a novel parallel semantic branch to precisely label different coral entities and delineate their boundaries. Trained on the CoralMask dataset, which includes over 41,000 labelled images, the technology enhances coral analysis by providing high-quality, detailed coral masks, improving scalability and accuracy in reef monitoring.

 

Dubai Future Solutions - Prototypes for Humanity engages an unmatched network of academic talent, including 20,000 professors from 800+ universities across 100+ countries, including 90% of the world’s top universities. The recognition by Dubai Future Solutions - Prototypes for Humanity adds to the accolades received by Prof. Yeung and his research team in IEEE/CVF Computer Vision and Pattern Recognition 2024. 

 

Please join us in congratulating Prof. Yeung and Mr. Zheng!

 

Learn more about the project: https://www.prototypesforhumanity.com/project/coral-reef-segmentation/ 
 


SHARE
TAGS
Innovation
Research and Technology
Research