A graph-based explorer for generative art that uses Neo4j for relationship storage, D3.js for interactive visualization, and AI for automated attribute extraction. The goal is to create a living map of AI-generated images where artists can explore visual relationships and discover new creative directions.

Problem

Generative art collections grow quickly but lack meaningful organization. File-system folders and tags don’t capture the rich relationships between pieces — shared styles, color palettes, compositional patterns, or thematic connections.

Approach

GraphStarz models art pieces as nodes in a graph database, with AI-extracted attributes (style, mood, palette, composition) forming edges between related works. D3.js renders the graph as an interactive, explorable visualization where artists can navigate by visual similarity rather than arbitrary categories.

What’s Implemented

• Neo4j graph schema for art pieces, attributes, and relationships
• AI extraction pipeline for automated attribute tagging of uploaded images
• D3.js visualization with interactive graph exploration
• Filtering and search across graph properties
• Whitelisted early access for beta testers

What This Demonstrates

Graph database design and querying, data visualization with D3.js, AI integration for automated content analysis, and full-stack development from database schema to interactive frontend.