GMR

Open source

[ICRA 2026] GMR: General Motion Retargeting. Retarget human motions into diverse humanoid robots in real time on CPU. Retargeter for TWIST.

2.5K stars0 likes0 views

Website GitHub

About

[ICRA 2026] GMR: General Motion Retargeting. Retarget human motions into diverse humanoid robots in real time on CPU. Retargeter for TWIST.

GMR: General Motion Retargeting

Key features of GMR:

Real-time high-quality retargeting, unlock the potential of real-time whole-body teleoperation, i.e., TWIST.
Carefully tuned for good performance of RL tracking policies.
Support multiple humanoid robots and multiple human motion data formats (See our table below).

[!NOTE] If you want this repo to support a new robot or a new human motion data format, send the robot files (.xml, .urdf, and meshes) / human motion data to Yanjie Ze or create an issue, we will support it as soon as possible. And please make sure the robot files you sent can be open-sourced in this repo.

This repo is licensed under the MIT License.

News & Updates

2026-01-21: GMR now supports Xsens BVH offline data.
2026-01-12: GMR now supports Fourier GR3, the 17th humanoid robot in the repo.
2025-12-02: GMR now supports TWIST2, which utilizes XRoboToolkit SDK.
2025-11-17: To join our community for discussions, you can add my WeChat contact QR Code with info like "[GMR] [Your Name] [Your Affiliation]".
2025-11-08: [MimicKit] from Jason Peng now supports GMR format. Check here.
2025-10-15: Now supporting PAL Robotics' Talos, the 15th humanoid robot.
2025-10-14: GMR now supports Nokov BVH data.
2025-10-14: Add a doc on ik config. See DOC.md
2025-10-09: Check TWIST open-sourced code for RL motion tracking.
2025-10-02: Tech report for GMR is now on arXiv.
2025-10-01: GMR now supports converting GMR pickle files to CSV (for beyondmimic), check scripts/batch_gmr_pkl_to_csv.py.
2025-09-25: An introduction on GMR is available on Bilibili.
2025-09-16: GMR now supports to use GVHMR for extracting human pose from monocular video and retargeting to robot.
2025-09-12: GMR now supports Tienkung, the 14th humanoid robot in the repo.
2025-08-30: GMR now supports Unitree H1 2 and PND Adam Lite, the 12th and 13th humanoid robots in the repo.
2025-08-28: GMR now supports Booster T1 for both 23dof and 29dof.
2025-08-28: GMR now supports using exported offline FBX motion data from OptiTrack.
2025-08-27: GMR now supports Berkeley Humanoid Lite, the 11th humanoid robot in the repo.
2025-08-24: GMR now supports Unitree H1, the 10th humanoid robot in the repo.
2025-08-24: GMR now supports velocity limits for the robot motors, use_velocity_limit=True by default in GeneralMotionRetargeting class (and we use 3*pi as the velocity limit by default); we also add printing of robot DoF/Body/Motor names and their IDs by default, and you can access them via robot_dof_names, robot_body_names, and robot_motor_names attributes.
2025-08-10: GMR now supports Booster K1, the 9th robot in the repo.
2025-08-09: GMR now supports Unitree G1 with Dex31 hands.
2025-08-07: GMR now supports Galexea R1 Pro (this is a wheeled humanoid robot!) and KUAVO, the 7th and 8th humanoid robots in the repo.
2025-08-06: GMR now supports HighTorque Hi, the 6th humanoid robot in the repo.
2025-08-04: Initial release of GMR. Check our twitter post.

Demos

Demo 1 Retargeting LAFAN1 dancing motion to 5 robots.	Demo 2 Galexea R1 Pro robot (view 1).	Demo 3 Galexea R1 Pro robot (view 2).	Demo 4 Switching robots by changing one argument.	Demo 5 HighTorque robot doing a twist dance.
Demo 6 Kuavo robot picking up a box.	Demo 7 Unitree H1 robot doing a ChaCha dance.	Demo 8 Booster T1 robot jumping (view 1).	Demo 9 Booster T1 robot jumping (view 2).	Demo 10 Unitree H1-2 robot jumping.
Demo 11 PND Adam Lite robot.	Demo 12 Tienkung robot walking.	Demo 13 Extracting human pose (GVHMR + GMR). ▶ Watch on Bilibili	Demo 14 PAL Robotics’ Talos robot fighting.	Demo 15 (Optional placeholder if you add a new one later!) Coming soon...

Supported Robots and Data Formats

Assigned ID	Robot/Data Format	Robot DoF	SMPLX (AMASS, OMOMO)	BVH LAFAN1	FBX (OptiTrack)	BVH Nokov	PICO (XRoboToolkit)
0	Unitree G1 `unitree_g1`	Leg (26) + Waist (3) + Arm (27) = 29	✅	✅	✅	✅	✅
1	Unitree G1 with Hands `unitree_g1_with_hands`	Leg (26) + Waist (3) + Arm (27) + Hand (2*7) = 43	✅	✅	✅	TBD	TBD
2	Unitree H1 `unitree_h1`	Leg (25) + Waist (1) + Arm (24) = 19	✅	TBD	TBD	TBD	TBD
3	Unitree H1 2 `unitree_h1_2`	Leg (26) + Waist (1) + Arm (27) = 27	✅	TBD	TBD	TBD	TBD
4	Booster T1 `booster_t1`	TBD	✅	TBD	TBD	TBD
5	Booster T1 29dof `booster_t1_29dof`	TBD	✅	✅	TBD	TBD
6	Booster K1 `booster_k1`	Neck (2) + Arm (24) + Leg (26) = 22	✅	TBD	TBD	TBD
7	Stanford ToddlerBot `stanford_toddy`	TBD	✅	✅	TBD	TBD
8	Fourier N1 `fourier_n1`	TBD	✅	✅	TBD	TBD
9	ENGINEAI PM01 `engineai_pm01`	TBD	✅	✅	TBD	TBD
10	HighTorque Hi `hightorque_hi`	Head (2) + Arm (25) + Waist (1) + Leg (26) = 25	✅	TBD	TBD	TBD
11	Galaxea R1 Pro `galaxea_r1pro` (this is a wheeled robot!)	Base (6) + Torso (4) + Arm (2*7) = 24	✅	TBD	TBD	TBD
12	Kuavo `kuavo_s45`	Head (2) + Arm (27) + Leg (26) = 28	✅	TBD	TBD	TBD
13	Berkeley Humanoid Lite `berkeley_humanoid_lite` (need further tuning)	Leg (26) + Arm (25) = 22	✅	TBD	TBD	TBD
14	PND Adam Lite `pnd_adam_lite`	Leg (26) + Waist (3) + Arm (25) = 25	✅	TBD	TBD	TBD
15	Tienkung `tienkung`	Leg (26) + Arm (24) = 20	✅	TBD	TBD	TBD
16	PAL Robotics' Talos `pal_talos`	Head (2) + Arm (27) + Waist (2) + Leg (26) = 30	✅	TBD	TBD	TBD
17	Fourier GR3 `fourier_gr3`	Head (2) + Arm (27) + Waist (3) + Leg (26) = 31	✅	TBD	TBD	TBD
More robots coming soon !
18	AgiBot A2 `agibot_a2`	TBD	TBD	TBD	TBD	TBD
19	OpenLoong `openloong`	TBD	TBD	TBD	TBD	TBD

Installation

[!NOTE] The code is tested on Ubuntu 22.04/20.04.

First create your conda environment:

bash

conda create -n gmr python=3.10 -y
conda activate gmr

Then, install GMR:

bash

pip install -e .

After installing SMPLX, change ext in smplx/body_models.py from npz to pkl if you are using SMPL-X pkl files.

And to resolve some possible rendering issues:

bash

conda install -c conda-forge libstdcxx-ng -y

Data Preparation

[SMPLX body model] download SMPL-X body models to assets/body_models from SMPL-X and then structure as follows:

bash

- assets/body_models/smplx/
-- SMPLX_NEUTRAL.pkl
-- SMPLX_FEMALE.pkl
-- SMPLX_MALE.pkl

[AMASS motion data] download raw SMPL-X data to any folder you want from AMASS. NOTE: Do not download SMPL+H data.

[OMOMO motion data] download raw OMOMO data to any folder you want from this google drive file. And process the data into the SMPL-X format using scripts/convert_omomo_to_smplx.py.

[LAFAN1