# TMElyralab/MuseV

MuseV: Infinite-length and High Fidelity Virtual Human Video Generation with Visual Conditioned Parallel Denoising

Repository: https://github.com/TMElyralab/MuseV
Canonical: https://ross.abutalabs.com/products/musev
Language: Python
License: NOASSERTION
License Family: other
Topics: diffusion, human-video-generation, image2video, video-generation, infinite-length, musev
Last push: 2024-06-28T04:21:48+00:00

## Health v2 (maintenance only)
Score: 25/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 63
- inputs: {"age_days": 891, "days_push": 796, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2846, forks 303 (observed 2026-08-28T04:07:25.007698+00:00)

## What it is
MuseV is a diffusion-based framework for generating high-fidelity virtual human videos of infinite length using a Visual Conditioned Parallel Denoising scheme. It supports Image2Video, Text2Image2Video, and Video2Video generation and is compatible with the Stable Diffusion ecosystem including LoRA, ControlNet, and IP-Adapter.

## Use cases
- generate videos of virtual humans from a single reference image
- create infinite-length talking avatar videos
- convert text prompts into human video clips
- restyle existing videos with reference images (video2video)
- build virtual human pipelines with MuseTalk and MusePose
- run a gradio demo to generate videos in a browser

## When to choose
- you need diffusion-based image-to-video generation of human characters
- you want long or infinite-length video generation beyond typical clip limits
- you want compatibility with Stable Diffusion checkpoints, LoRA, and ControlNet
- you are building a virtual human / digital avatar generation stack

## When to avoid
- you need real-time lip sync alone - use MuseTalk instead
- you need pose-controlled animation - use MusePose instead
- you lack a GPU or cannot run large diffusion models locally
- you need a production-ready, fully documented training pipeline - training code was not yet released

## Facets
- artifact type: library
- maturity: active
- function: video-processing, image-processing, machine-learning, deep-learning, stable-diffusion
- domain: artificial-intelligence, deep-learning, image-processing, computer-vision
- platform: python
- tags: diffusion-models, image-to-video, text-to-video, video-generation, virtual-human, parallel-denoising, infinite-length, generative-ai, video, linux, gpu, docker

## Member repositories
- TMElyralab/MuseV (main) score 25

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:25.007698+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 2026-08-30T07:37:01.668279+00:00, confidence not recorded.
  - readme: https://github.com/TMElyralab/MuseV (fetched 2026-08-28T04:07:25.007698+00:00, sha f6ff7126f0b6)
- Data as of 2026-08-30T08:39:29.467469+00:00.