EasyLive2D/relive2d

Live2D Library for Python (C++ impl): Supports model loading, lip-sync, basic face rigging, and precise click test.

What it solves

This project provides a way to load and operate Live2D models directly within Python, bypassing the need for Web Engines or indirect rendering methods. It allows developers to integrate high-quality 2D animated characters into Python-based applications using OpenGL.

How it works

The library is built as a Python C Extension that wraps the Live2D Native SDK (C++). By configuring an OpenGL context, the models can be rendered in any OpenGL-compatible window. It supports multiple versions of the Cubism SDK, providing separate modules for Cubism 2.1 (via pure Python or C++ ports) and Cubism 3.0+ (via C Extension).

Who it’s for

Developers creating virtual avatars, interactive 2D characters, or applications that require Live2D animations integrated into a Python environment (e.g., using Pygame, PyQt, or GLFW).

Highlights

  • Broad Compatibility: Supports Cubism 2.1 and 3.0+ models.
  • Direct Rendering: Uses OpenGL for high-performance rendering without a web browser wrapper.
  • Interactive Features: Includes support for gaze tracking, click interaction, lip-syncing, and precise part-based click detection.
  • UI Library Integration: Compatible with various UI libraries including Pygame, PyQt5, PySide, and GLFW.
  • Fine-grained Control: Allows control over model parameters, part transparency, and motion playback callbacks.

Related

  • Project
    playcanvas/supersplat-viewer

    A high-performance web viewer for Gaussian Splatting scenes, allowing developers to embed interactive 3D captures into websites using WebGPU or WebGL.

  • Project
    google/cameratrapai

    An ensemble of AI models that automates wildlife species classification in camera trap images using an object detector and a species classifier.

  • Project
    PozzettiAndrea/ComfyUI-SAM3

    A ComfyUI integration for Meta's SAM3 that enables open-vocabulary image and video segmentation using natural language text prompts.

  • Project
    naver/anny

    Anny is an open‑source, PyTorch‑based differentiable human‑body mesh model covering all ages. It offers multiple rigs (compact 104‑bone default, full MakeHuman), several mesh topologies (including SMPL‑X and SOMA), and fine‑grained shape, local‑change, and facial‑action parameters. Install via pip, run a quick Python example to output a .ply mesh, or launch an interactive Gradio demo. The library is Apache‑2.0 licensed, builds on MakeHuman assets, and is intended for research/production tasks such as pose estimation, animation, and cross‑age body modeling.