MarineGPT: MLLM for marine visual analysis
Project Description
MarineGPT is the first vision-language model specially designed for the marine domain, unlocking the secrets of the ocean to the public. MarineGPT not only pushes the boundaries of marine understanding to the general public but also offers a standard protocol for adapting a general-purpose assistant to downstream domain-specific experts.
Supervisor
YEUNG, Sai Kit
Quota
5
Course type
UROP1000
UROP1100
UROP2100
UROP3100
UROP3200
UROP4100
Applicant's Roles
Learn and implement MLLMs. Data collection and cleaning.
Applicant's Learning Objectives
Students will learn the basics of GPT-4V, Vision-Language Models (VLMs) and the evaluation metrics of evaluating domain-specific VLMs.
Complexity of the project
Moderate