# magic-research/magic-animate

[CVPR 2024] Official repository for "MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model"

Repository: https://github.com/magic-research/magic-animate
Canonical: https://ross.abutalabs.com/products/magic-animate
Homepage: https://showlab.github.io/magicanimate/
Language: Python
License: BSD-3-Clause
License Family: permissive
Last push: 2025-08-29T03:12:58+00:00

## Health v2 (maintenance only)
Score: 44/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 39, release rhythm 35, longevity 72
- inputs: {"age_days": 1016, "days_push": 369, "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 10897, forks 1081 (observed 2026-08-28T04:10:44.618286+00:00)

## What it is
MagicAnimate is the official implementation of a CVPR 2024 diffusion-based human image animation framework that animates a reference image following motion sequences from a driving video with temporal consistency. It builds on Stable Diffusion 1.5 with a DensePose-based ControlNet, appearance encoder, and temporal attention modules, and ships with inference code and a Gradio demo.

## Use cases
- animate a person photo using motion from a video
- transfer dance moves from a video onto a reference image
- animate oil paintings or movie characters performing actions
- animate multiple people following a given motion sequence
- combine text-to-image generation with human animation
- cross-identity animation using motion sequences from different people

## When to choose
- you need research-grade human image animation with temporal consistency
- you want to animate stylized or out-of-domain images like paintings
- you want a reproducible CVPR 2024 baseline for pose-driven video generation
- you have GPU resources and want local inference or a Gradio demo

## When to avoid
- you need real-time animation on consumer hardware
- you want a polished end-user product rather than research code
- you cannot download multi-gigabyte Stable Diffusion checkpoints
- you need active feature development or commercial support

## Facets
- artifact type: library
- maturity: maintenance
- function: video-processing, image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, artificial-intelligence, deep-learning
- platform: python
- tags: diffusion-model, human-animation, pose-transfer, stable-diffusion, temporal-consistency, cvpr-2024, gradio-demo, video, gpu, linux

## Member repositories
- magic-research/magic-animate (main) score 44

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:44.618286+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 2026-08-29T17:17:25.975924+00:00, confidence not recorded.
  - readme: https://github.com/magic-research/magic-animate (fetched 2026-08-28T04:10:44.618286+00:00, sha 193c14e1a487)
  - homepage: https://showlab.github.io/magicanimate/ (fetched 2026-08-29T08:16:18.643763+00:00, sha 3acb1060370d)
- Data as of 2026-08-30T08:39:29.467469+00:00.