Sea, Sense, and Melody - 3D Generation Algorithms
Project Description
The proposed project, Sea, Senses, and Melody, aims to create an environment-driven, generative, and interactive art experience related to HKUST’s natural surroundings. The goal is to encourage emotional sharing and empathy. The project will leverage a 3D virtual ecosystem where human users, as players, can engage in conversations with virtual beings driven by non-human data related to the sea near and around the campus.

In this experience, multi-sensory input from sensors placed on the HKUST main campus atrium balcony will drive an AI agent with its own virtual personality. This AI will generate moody poems and respond to users' emotional expressions, generate unique flowers and melody as response, which will be displayed on a large LED screen on campus. Instead of directly display user’s emotional expression, the proposed framework deals with it poetically and ambiguously, making the emotion sharing experience into a co-creation process between AI, environment and human.

This research aims to adapts a research-through-design, especially digital art-making.It is within the intersection of Human-Computer-Interaction(HCI), Interactive Digital artwork, social context of sustainability and emotional care, and humanity context of more-than-human. The applicant can learn and gain the ability to create interactive art using emerging AI technology, reflecting from a humanistic perspective on how art intervenes in public spaces. How can we use technological means to achieve AI, human, and environment co-created art that traditional methods cannot create.
Supervisor
BRAUD, Tristan Camille
Quota
2
Course type
UROP1000
UROP1100
UROP2100
UROP3100
UROP3200
UROP4100
Applicant's Roles
Develop the key algorithms and training framework to generate 3D flower models based on textual description. Applicant should be familiar with the architecture of 3D AI generation and have a foundation in training 3D AI.
Applicant's Learning Objectives
- Development experience in 3D model generation.
-Research existing models and understand front-end AIGC technologies. Through independent study of academic papers or open-source AIGC projects, propose a novel technical framework in the field of computer graphics.
-Understand and practice how to apply AI models in actual interactive art projects.
Complexity of the project
Moderate