Inspiration
I wanted to reimagine how people explore the world digitally. Maps and videos often feel flat and disconnected, so I asked myself: what if discovery could feel alive? The idea of a holographic mascot that not only shows places but also narrates their stories inspired the creation of Celestia AI.
What it does
Celestia AI transforms a map into an interactive 3D world where exploration feels personal. By dropping a marker on any location, a holographic mascot appears, presenting:
- Cinematic looping video snippets
- Six engaging fact cards per location
- Live voice narration
Instead of reading static text, users get a guided, immersive storytelling experience that makes culture, heritage, and history feel tangible.
How I built it
I built Celestia AI using a combination of frameworks and APIs:
- React for the UI framework
- Three.js/WebGL for rendering the 3D world and holographic effects
- Speech Synthesis API for voice narration of the fact cards
- Geolocation / Mapping APIs (OpenStreetMap + Nominatim) for handling location search and marker placement
- PEXELS APIs for fetching and embedding short looping video snippets tied to locations
- Custom synchronization logic to keep narration, fact cards, and video playback aligned in real time
Challenges I ran into
- Balancing performance with video overlays and 3D rendering
- Syncing narration with fact cards in real time
- Designing a futuristic yet usable interface
- Handling WebGL context loss during rendering
Accomplishments that I'm proud of
- Creating a working holographic explorer that blends video, narration, and facts seamlessly
- Designing a mascot system that feels engaging and interactive
- Building an immersive map interface beyond just navigation
What I learned
- How to combine 3D rendering, speech synthesis, and media synchronization into one flow
- The importance of crafting not just features, but a storytelling experience
- Handling advanced WebGL challenges to keep the interaction smooth
What's next for Celestia AI
- Adding mascots with unique personalities and voices
- Integrating real-time cultural event updates and local highlights
- Bringing in AR/VR support for a truly holographic experience
- Enhancing with adaptive background music that matches the mood of the narration
Built With
- git
- github
- groq
- javascript
- nominatim
- openstreetrmap
- pexelsapi
- react
- speech
- synthesis
- three.js
- typescript
- vercel
- webgl
Log in or sign up for Devpost to join the conversation.