Skip to content

Latest commit

 

History

History
30 lines (24 loc) · 1 KB

README.md

File metadata and controls

30 lines (24 loc) · 1 KB

simple_bodypix_python

A simple and minimal bodypix inference in python

Getting Started

  1. Install requirements
  2. Download tfjs models for bodypix.
# For example, download a ResNet50-based model to ./bodypix_resnet50_float_model-stride16
$ ./get-model.sh bodypix/resnet50/float/model-stride16
  1. Set path to models and image for inference in .py files
imagePath = './awesome_img.jpg'
modelPath = './bodypix_resnet50_float_model-stride16/'
  1. python3 evalbody_singleposemodel.py (Image with single person)

Observed Results

SINGLE POSE OUTPUT SinglePose Output SinglePose Part Heatmaps

Acknowledgement

  1. https://github.com/patlevin for support functions
  2. https://github.com/likeablob for download script

Note:

Multipose is work in progress. Pull Requests are welcome.