VNect tensorflow
This project is the tensorflow implementation of [VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera](http://gvv.mpi-inf.mpg.de/projects/VNect/), SIGGRAPH 2017. The project is written primarily in Python, distributed under the Apache License 2.0 license, first published in 2017. Key topics include: 3d-human-pose, depth-estimation, human-pose-estimation, siggraph, tensorflow.
VNect -- Tensorflow version
This project is the tensorflow implementation of VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera, SIGGRAPH 2017.
This is not an official implementation. Please contact paper author for related model.
Environments
- Ubuntu 16.04
- Python 2.7
- Tensorflow 1.3.0
- OpenCV 3.3.0
- OpenGL (optional)
Inference
- 1.Download model, put them in folder
models/weights - 2.Edit demo settings in shell script,
--device--demo_type--model_file--test_img--plot_2d--plot_3d - 3.If you have OpenGL, you can run
run_demo_tf_gl.shfor faster rendering of 3d joints. Otherwise, runrun_demo_tf.sh
TODO
- Some bugs in detected 3D joint locations.
- Training part of model.
Contributors
Showing top 3 contributors by commit count.
Related Repositories
mkocabas/VIBE
Official implementation of CVPR2020 paper "VIBE: Video Inference for Human Body Pose and Shape Estimation"
cbsudux/awesome-human-pose-estimation
A collection of awesome resources in Human Pose estimation.
wangzheallen/awesome-human-pose-estimation
Human Pose Estimation Related Publication
mks0601/3DMPPE_POSENET_RELEASE
Official PyTorch implementation of "Camera Distance-aware Top-down Approach for 3D Multi-person Pose Estimation from a Single RGB Image", ICCV 2019
hongsukchoi/Pose2Mesh_RELEASE
Official Pytorch implementation of "Pose2Mesh: Graph Convolutional Network for 3D Human Pose and Mesh Recovery from a 2D Human Pose", ECCV 2020
Daniil-Osokin/lightweight-human-pose-estimation-3d-demo.pytorch
Real-time 3D multi-person pose estimation demo in PyTorch. OpenVINO backend can be used for fast inference on CPU.