shinshin86/mesh-avatar-studio

Turn one illustration into an animated 2D mesh avatar with a coding agent and a local editor

What it solves

This project simplifies the creation of animated 2D avatars from static illustrations. It automates the complex process of rigging, layering, and mesh deformation, allowing users to turn a single image into a character that can blink, talk, sway, and track facial movements for streaming.

How it works

The workflow combines AI-assisted automation with manual fine-tuning. First, a coding agent (such as Claude Code or Codex) processes an uploaded image to create the initial rig, layer cuts, and mesh structure. Users then open the project in a local editor to manually adjust the mesh points, refine layer masks, and generate specific mouth or eye variants. The final avatar can be exported as a single file and used in a browser-based live-streaming interface that supports real-time face tracking and lip-syncing via MediaPipe.

Who it’s for

It is designed for streamers, VTubers, and digital creators who want to create custom animated avatars without needing advanced knowledge of 3D modeling or complex animation software.

Highlights

  • Uses AI coding agents to automate the tedious initial setup of avatar rigging and layering.
  • Provides a local, browser-based editor for precise manual adjustments and live previews.
  • Supports real-time face tracking and microphone-based lip-syncing for streaming platforms like OBS.
  • Includes an exportable .mavatar format that bundles all rig data, layers, and expressions.
  • Offers a zero-install player for testing and using avatars directly in the browser.

Related

  • Project
    ButzYung/SystemAnimatorOnline

    A webcam-based AI motion capture application for animating 3D avatars (VRM/MMD) in real-time for VTubing and XR purposes.

  • Project
    JohnHeibel/ClaudeAnimationBase

    A starter kit for animating characters using AI coding agents, combining p5.js and p5.brush to generate stylized 2D videos from text prompts.

  • Project
    xikhar/persona

    A cross-platform desktop application that provides a real-time, expressive 3D character avatar that reacts to voice output from AI assistants.

  • Project
    jaredrhod/barehands

    A webcam-based hand-tracking interface that lets you manipulate notes, images, and 3D models spatially, designed to be wired into an AI agent as a visual and interactive body.

  • Project
    CookSleep/gpt_image_playground

    A feature-rich web and desktop interface for managing AI image generation and editing, supporting multiple providers, batch processing, and local-first history management.