# lizhe00/AnimatableGaussians

Code of [CVPR 2024] "Animatable Gaussians: Learning Pose-dependent Gaussian Maps for High-fidelity Human Avatar Modeling"

Repository: https://github.com/lizhe00/AnimatableGaussians
Canonical: https://ross.abutalabs.com/products/animatablegaussians
Homepage: https://animatable-gaussians.github.io/
Language: Python
License: NOASSERTION
License Family: other
Topics: 3d-human, 3d-reconstruction, animatable-avatar, 3d-gaussian-splatting
Last push: 2024-11-16T13:23:25+00:00

## Health v2 (maintenance only)
Score: 27/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 0, release rhythm 35, longevity 72
- inputs: {"age_days": 1013, "days_push": 655, "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 1093, forks 80 (observed 2026-08-28T04:03:33.685168+00:00)

## What it is
Official PyTorch implementation of the CVPR 2024 paper 'Animatable Gaussians', which learns pose-dependent Gaussian maps for high-fidelity animatable human avatar modeling from RGB videos. It combines 2D CNNs (StyleGAN-based) with 3D Gaussian splatting and a learned parametric template to create lifelike, animatable avatars.

## Use cases
- create animatable 3d human avatars from rgb videos
- reconstruct high-fidelity 3d humans with gaussian splatting
- animate avatars with novel poses from motion capture data
- research on pose-dependent human appearance modeling
- build digital humans for games or virtual production
- compare against nerf-based avatar methods

## When to choose
- you need photorealistic animatable human avatars from multi-view or rgb video
- you want to reproduce or extend a cvpr 2024 avatar modeling method
- you need pose-dependent garment detail that nerf-based methods struggle with
- you have gpu resources and smpl-based preprocessing pipelines available

## When to avoid
- you need a production-ready avatar solution with a polished api or gui
- you lack multi-view rgb video or the required smpl/actorthq preprocessing assets
- you need real-time avatar animation on consumer hardware without a gpu
- you want a general-purpose 3d reconstruction tool not specific to humans

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, graphics, simulation
- domain: computer-vision, graphics, machine-learning, deep-learning
- platform: python
- tags: 3d-gaussian-splatting, human-avatar, animatable-avatar, 3d-reconstruction, cvpr-2024, research-code, nerf-alternative, linux, gpu

## Member repositories
- lizhe00/AnimatableGaussians (main) score 27

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:33.685168+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 2026-08-30T06:48:57.513117+00:00, confidence not recorded.
  - readme: https://github.com/lizhe00/AnimatableGaussians (fetched 2026-08-28T04:03:33.685168+00:00, sha 2fbdc8d33986)
  - homepage: https://animatable-gaussians.github.io/ (fetched 2026-08-29T12:50:51.794088+00:00, sha 5dd711b6ab26)
- Data as of 2026-08-30T08:39:29.467469+00:00.