WayangDex is a Pokédex-inspired digital encyclopedia web application developed as an undergraduate thesis project. It leverages computer vision to help users identify, collect, and learn about traditional Indonesian shadow puppet (Wayang Kulit) characters in an interactive, gamified environment.
Powered by a custom YOLO model (exported to ONNX format) served via a Python FastAPI backend, WayangDex processes captured photos on-demand to identify characters, unlock detailed profiles, character lore, and progression badges.
- Event-Driven AI Detection: Server-side character identification triggered via photo capture using ONNX Runtime Python.
- Digital Encyclopedia & Lore: Comprehensive background info, character traits, alternate names, and audio narration.
- Gamified Collection System: Pokédex-style index featuring completion tracking across distinct character classes.
- Progression Dashboard: User leveling system, XP rewards, total scan counters, and unlockable achievement badges.
- Backend Framework: FastAPI (Python)
- Computer Vision Model: YOLO (Exported to
.onnx) - Inference Engine: ONNX Runtime (
onnxruntime) - Package & Env Manager:
uv - Frontend Architecture: HTML5 / Server-Side Rendered UI
- License: MIT License
This project is licensed under the MIT License - see the LICENSE file for details.





