# SeaSense
Public spaces rarely invite slow, shared reflection. The task was to turn ordinary text about feelings into a collective, ambient visualization people could contribute to on the spot.
Overview
SeaSense turns hundreds of private feelings into a living public artwork. Visitors open a lightweight app, type how they feel, and watch a language model interpret their words as an abstract flower. A custom 3D pipeline renders that bloom and plants it on a collective canvas projected across a video wall — each blossom unique, yet woven into a shared garden of campus emotion. The system stays ambiguous and respectful: no categories or emotion scores, just open-ended prose that the AI translates into colour, shape, and motion. The pipeline links GPT-4 flower descriptions with Hunyuan3D generation, post-processing, and a Unity installation that quietly refreshes as more people contribute. SeaSense debuted as a weeklong exhibition, drawing more than 300 passersby who paused to write notes, take photos of the evolving mural, and chat about how the collective mood shifted hour by hour. Two follow-up studies compared versions of the experience with and without AI co-creation, showing that the generative layer invites longer reflections and sparks more curiosity about others' emotions while keeping contributors comfortable.
Problem
Emotion tech often collapses nuance into fixed labels, discouraging participation and limiting personal resonance in public settings.
Approach
Use an LLM to interpret free-text emotions and drive a 3D flower generation pipeline in Unity. Add gentle guardrails for privacy/moderation and a phone-friendly input flow for walk-up participation.
Impact
- * Week-long deployment with 300+ contributions created a continuously evolving garden.
- * Follow-up interviews indicated increased curiosity and deeper reflective engagement.
- * Operational playbook for on-site setup, moderation, and teardown is now repeatable.