Immersive Gallery - AI Tinkerers - New York City Hackathon
AI Tinkerers - New York City
Hackathon Showcase 3rd Place Winner

Immersive Gallery

Build taste into multimodal models

4 members Watch Demo

Immersive Gallery is a dynamic UI for showcasing images in 3D, adaptive based on user input. Users can leverage the copilot chatbox to change the UI. e.g. We have emojis that display as you burst through images, an image upload capability to add it in real-time, adaptive color background for each image, background soundtrack that can be unmuted, a chatbox to enable new features such as adding a frame around images or to disable the existing features.

The interface pairs chat with traditional default layout settings. We want users to re-render new UIs based on the image through the chatbox, but to also navigate the gallery as is or for an exploration of existing functions on how to re-render it. Because the default UI is navigating through images in multiple planes that enables 3D navigation of flat 2D images, we believe the dynamic UI will enhance the users ability to explore and pick images. We used Anthropic Claude Code to develop it, and copilot Kit to help re-render the UI based on chat input.

https://github.com/edoardolunardi/infinite-canvas

Anthropic Betaworks CopilotKit