Flower Garden
2025, Installation & Creative Coding made with p5.js
Created for Arts for Chatbot’s Sake Taught by Professor Carrie Wang
Exhibited in NYU Spring Show 2025
Flower Garden is an interactive installation that combines natural language processing, emotion detection, and generative visuals. The system uses OpenAI's API to respond to user input in real time, automatically detecting the language being spoken to ensure a seamless experience across different users.
As the conversation unfolds, the chatbot analyzes the emotional content of each message and generates flowers and growth patterns accordingly — each emotion producing a distinct visual output. The flora is rendered using an L-system, a rule-based algorithmic structure commonly used to simulate natural growth, which gives each bloom a sense of organic logic.
The installation also incorporates hand detection. A butterfly tracks the user's hand in real time, following its position as it moves through the frame. When the hand approaches the generated flowers, the butterfly responds and interacts with them, creating a feedback loop between the user's physical presence and the virtual ecosystem on screen.
The research was divided along similar lines — I conducted qualitative research through user interviews, while Irene handled the quantitative side. Together, the two approaches gave us a more complete picture of how people actually experienced the piece.
Elevator Pitch Deck
NYU Spring Show 2025