smpl-x 开源项目推荐（35 个）

cuevhv /mamma

MAMMA：无标记精确多人运动采集的官方代码。

Rubikplayer /flame-fitting

Example code for the FLAME 3D head model. The code demonstrates how to sample 3D heads from the model, fit the model to 3D keypoints and 3D scans.

squall01337 /mixamo-llm-mocap

将任何视频转换为Mixamo-rig动画- GVHMR估计器、规格驱动的重定向、通过HCP在Blender中应用FK。可与任何Mixamo角色配合使用;专为由AI代理端到端操作而设计。

naver /multi-hmr

Pytorch demo code and models for Multi-HMR

openxrlab /xrmocap

OpenXRLab Multi-view Motion Capture Toolbox and Benchmark

muelea /shapy

CVPR 2022 - Official code repository for the paper: Accurate 3D Body Shape Regression using Metric and Semantic Attributes.

DavidBoja /SMPL-Anthropometry

测量SMPL体模型

isarandi /smplfitter

The Fast Way From Vertices to Parametric 3D Humans

ZhengdiYu /SignAvatars

(ECCV 2024) SignAvatars：大规模 3D 手语整体运动数据集与基准

PerceivingSystems /bedlam_render

BEDLAM (CVPR 2023) render pipeline tools

nghorbani /homogenus

Human Image Gender Classifier for Expressive Body Capture

Xiangyue-Zhang /SemTalk

🔥 [ICCV 2025] SemTalk: Holistic Co-speech Motion Generation with Frame-level Semantic Emphasis

twehrbein /ExMR

3D body, hand and foot motion (SMPL-X) from a single monocular video: world-grounded, with export for Blender, Unity, Unreal and humanoid robots. Official code of ExMR (arXiv 2610.26213).

KyujinHan /Smplify-X-Perfect-Implementation

Smplify-X implementation. (2025. 04. 07 No Error & Recent version)

Xiangyue-Zhang /StreamTalk

🔥 [ECCV 2026] StreamTalk: Streaming Co-Speech Gesture Generation with Key-Pose Anchoring

Xiangyue-Zhang /EchoMask

[🔥ACM MM2025]EchoMask：基于注意力的语音查询模拟，用于整体共言动作生成

PerceivingSystems /bedlam2_render

BEDLAM 2.0 (NeurIPS 2025) render pipeline tools

darcyjia /ue5-ai-mocap

AI 视频 → 双人动捕 → UE5 MetaHuman：Seedance 生成武打视频，GVHMR 解算，脚部锁定，手型 / 受力 / 表情，港片风格运镜预演，一键渲染。UE 5.8 编辑器插件 + 命令行。

PerceivingSystems /bedlam2_retargeting

BEDLAM 2.0 SMPL-X motion retargeting tool using Unreal Engine’s IK Retargeter.

Xiangyue-Zhang /GlobalDiff

🔥 [AAAI 2026] Mitigating Error Accumulation in Co-Speech Motion Generation via Global Rotation Diffusion and Multi-Level Constraints