Skip to content

Latest commit

 

History

History
25 lines (19 loc) · 1.37 KB

README.md

File metadata and controls

25 lines (19 loc) · 1.37 KB
Raw

simple_posenet_python

A simple and minimal posenet inference in python

Getting Started

  1. Install requirements
  2. Download tfjs models for posenet.
  3. Set path to models and image for inference in .py files
  4. python3 eval_singleposemodel.py (Image with single person) or python3 eval_multiposemodel.py (Image with single/multiple person)

Observed Results

SINGLE POSE OUTPUT SinglePose BBOX SinglePose Keypoints SinglePose Connected Keypoints

MULTIPOSE OUTPUT MultiPose BBOX MultiPose Keypoints MultiPose Connected Keypoints

Multipose gives better results than singlepose even for single person. Due to graph tree based refinement.

ACKNOWLEDGEMENT

Thanks to https://github.com/patlevin for support functions