# menyifang/MIMO

Official implementation of "MIMO: Controllable Character Video Synthesis with Spatial Decomposed Modeling"

Repository: https://github.com/menyifang/MIMO
Canonical: https://ross.abutalabs.com/products/menyifang-mimo
Language: Python
License: Apache-2.0
License Family: permissive
Topics: character-animation, diffusion-models, human-animation, image-animation, video-editing, video-generation
Last push: 2025-06-19T01:52:26+00:00

## Health v2 (maintenance only)
Score: 34/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 27, release rhythm 35, longevity 50
- inputs: {"age_days": 708, "days_push": 441, "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 1578, forks 72 (observed 2026-08-28T04:05:06.267502+00:00)

## What it is
MIMO is the official PyTorch implementation of a CVPR 2025 paper on controllable character video synthesis using spatially decomposed modeling with diffusion models. It animates a single character image using driving 3D poses or in-the-wild videos, with controllable character, motion, and scene attributes.

## Use cases
- animate a character image with 3D pose sequences from motion datasets
- generate character videos driven by in-the-wild videos
- control character appearance, motion, and scene independently in video synthesis
- create talking or moving avatar videos from a single photo
- reproduce research results for controllable video synthesis

## When to choose
- you need research-grade controllable character video generation from a single image
- you want to animate characters with 3D pose or video drivers
- you have access to high-end GPUs like A100 or L20

## When to avoid
- you lack a CUDA GPU with substantial VRAM
- you need a production-ready end-user application rather than research code
- you need real-time performance on consumer hardware

## Facets
- artifact type: library
- maturity: active
- function: video-processing, image-processing, machine-learning, deep-learning
- domain: computer-vision, artificial-intelligence, deep-learning
- platform: python
- tags: diffusion-models, character-animation, human-animation, video-generation, pose-driven-animation, research-code, cvpr-2025, video, gpu, linux

## Member repositories
- menyifang/MIMO (main) score 34

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:06.267502+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 2026-08-30T03:56:55.980501+00:00, confidence not recorded.
  - readme: https://github.com/menyifang/MIMO (fetched 2026-08-28T04:05:06.267502+00:00, sha 0c53e957a3a4)
- Data as of 2026-08-30T08:39:29.467469+00:00.