# yuyuyzl/EasyVtuber

Based on Talking-head-anime 3, works like Vtube Studio.

Repository: https://github.com/yuyuyzl/EasyVtuber
Canonical: https://ross.abutalabs.com/products/easyvtuber
Language: Python
License: MIT
License Family: permissive
Last push: 2026-02-12T07:11:57+00:00

## Health v2 (maintenance only)
Score: 62/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 67, release rhythm 35, longevity 100
- inputs: {"age_days": 1593, "days_push": 202, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3051, forks 272 (observed 2026-08-28T04:07:39.646395+00:00)

## What it is
EasyVtuber is a Python-based VTubing application built on the Talking Head Anime model that turns a single anime character illustration into a live-animated avatar driven by face tracking. It supports iPhone TrueDepth (iFacialMocap) and webcam-based tracking, GPU-accelerated inference (TensorRT/ONNX/DirectML), frame interpolation, super-resolution, and transparent-channel output to OBS via Spout2 or virtual camera.

## Use cases
- animate an anime character with face tracking for streaming
- vtubing without buying a Live2D model
- use iPhone face capture to drive an anime avatar in OBS
- run a talking-head anime avatar on AMD or Intel GPUs
- stream a virtual avatar with transparent background in OBS
- webcam-based face mocap for a 2D anime character

## When to choose
- you want a VTube Studio-like experience using a single drawn character image instead of a rigged Live2D model
- you have a gaming GPU (NVIDIA, AMD, or Intel) and want local, real-time avatar animation
- you want high-refresh iPhone ARKit-style face tracking over UDP without extra PC middleware
- you need alpha-channel avatar output composited directly into OBS

## When to avoid
- you need a full Live2D rig with custom physics and per-part animation control
- you are on Linux or macOS, since the solution is tested on Windows 10/11
- you have no discrete or capable integrated GPU, as inference is GPU-heavy
- you want a plug-and-play mobile app rather than a locally installed Python toolchain

## Facets
- artifact type: application
- maturity: active
- function: computer-vision, machine-learning, image-processing, video-processing, graphics, gui
- domain: computer-vision, image-processing, graphics, entertainment, machine-learning
- platform: windows, python
- tags: vtuber, face-tracking, talking-head-anime, obs-virtual-camera, ifacialmocap, tensorrt, onnx, directml, spout2, frame-interpolation, super-resolution, anime-avatar, streaming, real-time, gpu, desktop

## Member repositories
- yuyuyzl/EasyVtuber (main) score 62

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:39.646395+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:28:54.559681+00:00, confidence not recorded.
  - readme: https://github.com/yuyuyzl/EasyVtuber (fetched 2026-08-28T04:07:39.646395+00:00, sha c8f2164febc0)
- Data as of 2026-08-30T08:39:29.467469+00:00.