Skip to content

Latest commit

 

History

History
43 lines (30 loc) · 1.24 KB

README.md

File metadata and controls

43 lines (30 loc) · 1.24 KB
Raw

Face-Keypoints

How to Use

  • Clone this repo to your machine git clone https://github.com/Rajrup/Face-Keypoints.git
  • Install the python dependencies (see below)
  • Download the checkpoints(see below)
  • Double check the model paths
  • Ready to run pipeline.py

Checkpoint Preparation

  • Find all the models: Google Drive
  • Extract the Tensorflow model files into models/ folder.
    • Face: Path will look this - ./models/frozen_inference_graph_face.pb
    • PRNet: Path will look this - ./models/PRNet/net-data/256_256_resfcn256_weight.data-00000-of-00001

Requirements

This code has been tested in Python 3.7. See requirements.txt for python packages.

pip install -r requirements.txt

Running Pipeline

  • Tensorflow Pipeline:

    python pipeline.py

  • Tensorflow Serving Pipeline: TODO

  • Output in the media folder. Check the file output.avi for keypoints.

Components

One module's output will go to the next one

  • Video Reader
  • Face Detection (Face)
  • Key Point Extraction (PRNet)