AmiMoji is a web application that allows users to create and animate 3D avatars based on their facial movements. Users can capture these animations and save them as GIFs or MP4 videos.
- Frontend: Next.js, NextUI
- Backend: Firebase Database
- 3D Rendering: Three.js
- Facial Movement Recognition: Mediapipe
AmiMoji uses Mediapipe to track facial movements through the user's webcam. This data is then applied to a 3D avatar rendered with Three.js, allowing the avatar to mirror the user's expressions instantly.
Users can customize their avatars with different styles, colors, and accessories. The customization options include hairstyles, outfits, glasses, and more, providing a personalized experience.
Once satisfied with their avatar's animation, users can capture the movement and export it as either a GIF or an MP4 video. This feature is accessible directly from the interface, making it easy to share creations on social media or with friends.
Firebase is used for real-time data storage and synchronization. User captured animations are stored securely in Firebase Firestore.
The application is designed to be responsive, ensuring a consistent user experience across desktops, tablets, and mobile devices. The interface adapts dynamically to different screen sizes and orientations.
Recording
Exporting
Contributions are welcome! Please fork the repository, create a branch, and submit pull requests to contribute new features, improvements, or bug fixes.