Variational human pose prior for human pose synthesis and estimation.
Project description
VPoser: Variational Human Pose Prior
Description
Human joint configuration, also called as pose, is restricted by biomechanics of our body. Utilizing these constrains accuratly would be a corner stone of many computer vision tasks, such as estimating 3D human body parameters from 2D keypoints, detecting anomolies, and etc.
Here we present a method that is used in SMPLify-X. Our variational human pose prior, named as VPoser, has the following features:
- is end-to-end differentiable
- provides a way to penalizes impossible poses while allowing possible ones
- effectively considers interdependency of configurations of the joints
- intorduces a more efficient, and lower dimensional representation for human pose
- can be used as a generative source for data dependent tasks
Table of Contents
- Description
- Installation
- [Loading trained models](#Loading trained models)
- Example
- Citation
- Contact
- License
Installation
To install the model simply you can:
- To install from PyPi simply run:
pip install human_body_prior
- Clone this repository and install it using the setup.py script:
git clone https://github.com/nghorbani/human_body_prior
python setup.py install
Loading Trained Models
To download the VPoser trained models go to the project website and register to get access to the downloads section. Afterwards, you can follow model loading tutorial to load and use your trained models.
Train VPoser
We train VPoser, using a variational autoencoder, which learns a latent representation of human pose and regularizes the distribution of the latent code to be a normal distribution. We train our prior on the data released by AMASS, namely SMPL pose parameters of various publicly available human motion capture datasets. You can follow data preparation tutorial to learn how to download and prepare AMASS for VPoser. Afterwards, you can Train VPoser from scratch.
Tutorials
- VPoser pose space for SMPL body model
- Sampling novel poses from VPoser
- Preparing the training dataset
- Train VPoser from scratch
License
Software Copyright License for non-commercial scientific research purposes. Please read carefully the terms and conditions and any accompanying documentation before you download and/or use the SMPL-X/SMPLify-X model, data and software, (the "Model & Software"), including 3D meshes, blend weights, blend shapes, textures, software, scripts, and animations. By downloading and/or using the Model & Software (including downloading, cloning, installing, and any other use of this github repository), you acknowledge that you have read these terms and conditions, understand them, and agree to be bound by them. If you do not agree with these terms and conditions, you must not download and/or use the Model & Software. Any infringement of the terms of this agreement will automatically terminate your rights under this License.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for human_body_prior-0.8.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | d642b7eacf4ddb29015547f71fb52c4c14e91a1905d89f066be303166d2605a8 |
|
MD5 | 25fd29b6e246108c8c4d1486e0411558 |
|
BLAKE2b-256 | e9c4e2145f769645d5be9fb88abc4ee35c9725462da38c85ca567dfbd4faadf1 |