# EricGuo5513/momask-codes

Official implementation of "MoMask: Generative Masked Modeling of 3D Human Motions (CVPR2024)"

Repository: https://github.com/EricGuo5513/momask-codes
Canonical: https://ross.abutalabs.com/products/momask-codes
Homepage: https://ericguo5513.github.io/momask/
Language: Python
License: MIT
License Family: permissive
Topics: 3d-generation, animation, motion, motion-generation, motion-synthesis, text-driven, text-to-motion
Last push: 2024-09-13T19:09:40+00:00

## Health v2 (maintenance only)
Score: 27/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 72
- inputs: {"age_days": 1008, "days_push": 719, "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 1308, forks 116 (observed 2026-08-28T04:04:19.221842+00:00)

## What it is
Official PyTorch implementation of MoMask (CVPR 2024), a masked modeling framework that generates 3D human motion sequences from text prompts using hierarchical vector quantization and bidirectional transformers. It includes training, evaluation, and generation scripts, plus a Hugging Face demo and a Blender add-on.

## Use cases
- generate 3d human motion animations from text descriptions
- text-to-motion synthesis for characters
- temporal inpainting of motion sequences
- export bvh animations for blender or game engines
- research baseline for motion generation models
- evaluate text-driven motion generation on humanml3d

## When to choose
- you need state-of-the-art text-to-motion generation with pretrained models
- you want to reproduce or extend a published CVPR 2024 motion generation method
- you need BVH output for animation pipelines like Blender

## When to avoid
- you need production-ready software with long-term support
- you work outside Python/PyTorch environments
- you need real-time motion generation on low-end hardware without setup effort

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, animation, graphics, nlp
- domain: machine-learning, computer-vision, graphics, artificial-intelligence
- platform: python, cross-platform
- tags: text-to-motion, 3d-human-motion, masked-modeling, cvpr-2024, motion-synthesis, bvh, research-code, transformer, game-development, linux, gpu

## Member repositories
- EricGuo5513/momask-codes (main) score 27

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:19.221842+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 2026-08-30T04:50:02.633332+00:00, confidence not recorded.
  - readme: https://github.com/EricGuo5513/momask-codes (fetched 2026-08-28T04:04:19.221842+00:00, sha 4819d8e22cd0)
  - homepage: https://ericguo5513.github.io/momask/ (fetched 2026-08-29T12:08:19.148003+00:00, sha 1007d2e8e47f)
- Data as of 2026-08-30T08:39:29.467469+00:00.