fmpose3d 0.0.7

No project description provided

FMPose3D: monocular 3D pose estimation via flow matching

This is the official implementation of the approach described in the paper:

FMPose3D: monocular 3D Pose Estimation via Flow Matching
Ti Wang, Xiaohang Yu, Mackenzie Weygandt Mathis

🚀 TL;DR

FMPose3D replaces slow diffusion models for monocular 3D pose estimation with fast Flow Matching, generating multiple plausible 3D poses via an ODE in just a few steps, then aggregates them using a reprojection-based Bayesian module (RPEA) for accurate predictions, achieving state-of-the-art results on human and animal 3D pose benchmarks.

News!

• Feb 2026: FMPose3D code and arXiv paper is released - check out the demos here or on our project page
• Planned: This method will be integrated into DeepLabCut

Installation

Set up an environment

Make sure you have Python 3.10+. You can set this up with:

conda create -n fmpose_3d python=3.10
conda activate fmpose_3d

git clone https://github.com/AdaptiveMotorControlLab/FMPose3D.git
# TestPyPI (pre-release/testing build)
pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ fmpose3d==0.0.7
# Future Official PyPI release
# pip install fmpose3d

Demo

Testing on in-the-wild images (humans)

This visualization script is designed for single-frame based model, allowing you to easily run 3D human pose estimation on any single image.

Before testing, make sure you have the pre-trained model ready. You may either use the model trained by your own or download ours from here and place it in the ./pre_trained_models directory.

Next, put your test images into folder demo/images. Then run the visualization script:

sh vis_in_the_wild.sh

The predictions will be saved to folder demo/predictions.

Training and Inference

Dataset Setup

Setup from original source

You can obtain the Human3.6M dataset from the Human3.6M website, and then set it up using the instructions provided in VideoPose3D.

Setup from preprocessed dataset (Recommended)

You also can access the processed data by downloading it from here.

Place the downloaded files in the dataset/ folder of this project:

<project_root>/
├── dataset/
│   ├── data_3d_h36m.npz
│   ├── data_2d_h36m_gt.npz
│   └── data_2d_h36m_cpn_ft_h36m_dbb.npz

Training

The training logs, checkpoints, and related files of each training time will be saved in the './checkpoint' folder.

For training on Human3.6M:

sh /scripts/FMPose3D_train.sh

Inference

First, download the folder with pre-trained model from here and place it in the './pre_trained_models' directory.

To run inference on Human3.6M:

sh ./scripts/FMPose3D_test.sh

Experiments Animals

For animal training/testing and demo scripts, see animals/README.md.

Acknowledgements

We thank the Swiss National Science Foundation (SNSF Project # 320030-227871) and the Kavli Foundation for providing financial support for this project.

Our code is extended from the following repositories. We thank the authors for releasing the code.

• MHFormer
• StridedTransformer-Pose3D
• ST-GCN
• VideoPose3D