# Tencent/MimicMotion

High-Quality Human Motion Video Generation with Confidence-aware Pose Guidance

Repository: https://github.com/Tencent/MimicMotion
Canonical: https://ross.abutalabs.com/products/mimicmotion
Homepage: https://tencent.github.io/MimicMotion/
Language: Python
License: NOASSERTION
License Family: other
Topics: diffusion-models, video-generation
Last push: 2025-11-18T06:29:12+00:00

## Health v2 (maintenance only)
Score: 47/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 52, release rhythm 35, longevity 57
- inputs: {"age_days": 800, "days_push": 288, "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 2647, forks 240 (observed 2026-08-28T04:07:06.807520+00:00)

## What it is
MimicMotion is a diffusion-based framework from Tencent for generating high-quality human motion videos guided by pose sequences, featuring confidence-aware pose guidance and progressive latent fusion for long, temporally smooth videos. It ships as a Python research codebase with pretrained checkpoints published on Hugging Face.

## Use cases
- generate videos of a person dancing from a reference image and pose sequence
- animate a static character photo with human motion capture data
- create long, temporally smooth human motion videos with diffusion models
- research controllable pose-guided video generation
- compare pose-driven video generation baselines like MuseV and Magic Dance

## When to choose
- you need pose-conditioned human motion video generation with good temporal smoothness
- you want to generate videos of arbitrary length from a single reference image
- you are doing research on diffusion-based video generation and want an ICML-published baseline

## When to avoid
- you need general-purpose text-to-video generation without pose guidance
- you lack a GPU or cannot run heavy diffusion inference locally
- you need a production-ready end-user application rather than a research codebase

## Facets
- artifact type: library
- maturity: active
- function: video-processing, machine-learning, deep-learning, image-processing
- domain: deep-learning, computer-vision, artificial-intelligence
- platform: python
- tags: diffusion-models, pose-guided-video-generation, human-motion, video-generation, research, video, gpu, linux

## Member repositories
- Tencent/MimicMotion (main) score 47

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:06.807520+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 2026-08-30T02:18:56.755949+00:00, confidence not recorded.
  - readme: https://github.com/Tencent/MimicMotion (fetched 2026-08-28T04:07:06.807520+00:00, sha d679082963b1)
  - homepage: https://tencent.github.io/MimicMotion/ (fetched 2026-08-29T10:01:42.041614+00:00, sha 79e7c755d443)
- Data as of 2026-08-30T08:39:29.467469+00:00.