How about animated Memoji in Lottie?

Background

I couldn't find a way to embed animated stickers from iMessage into websites. So, I wrote my own solution. I hope it helps you too!

Introduction

This project allows you to convert animated Memoji from .mov format to Lottie. Lottie is a powerful animation format that is well-suited for web applications because it supports transparent backgrounds and is lightweight. The default export from iMessage is in .mov format, which not only has a dark background but is also not intended for direct use on websites. This project aims to provide a better alternative by converting those animations into the Lottie format.

Features

• Convert .mov animated Memoji to Lottie format.
• Remove dark backgrounds using the OpenCV library.
• Convert each frame to .webp format for efficient web use.
• Generate a Lottie JSON file ready for embedding in web projects.

Usage

1. Clone this repository:

   git clone https://github.com/machnevegor/memoji
   cd memoji

2. Install the dependencies using Poetry:

   poetry install

3. Create your animated Memoji in iMessage. For guidance, check out this support article.

4. Export the Memoji as a .mov file to the data/ directory.

5. Open the Jupyter Notebook notebook.ipynb.

6. Update the INPUT_VIDEO and OUTPUT_JSON paths in the notebook to point to your .mov file and desired output JSON file.

7. Run the cells in the notebook to process the animation.

8. Check the output: Review the generated JSON file to ensure it meets your quality standards.

9. Copy the generated JSON file to your website. Enjoy!

Important Notes

• The conversion process includes a compression step when generating .webp files. Please note that this process is not lossless, meaning some quality may be reduced. Make sure to review the output to ensure it meets your needs.
• If you encounter any issues or have questions, feel free to open an issue in this repository.

Contributing

Contributions are welcome! If you have suggestions or improvements, please create a pull request or open an issue.