FTracker

This face tracker is proposed in a bachelor thesis at HCM University of Science, Vietnam National University.

The distilled work is detailed in manuscript-1.pdf

Full documents can be found at: (Upcoming...)

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.

Prerequisites

This project require installing of these tools:

• Docker CE (~=18.09)
• NVidia Docker driver

Installing

This is a step by step guide to help you install this project

First, build the Docker image using this command

docker build . -f Dockerfile --tag face

Build a network dockernet

docker network create -d bridge --subnet 192.168.0.0/24 --gateway 192.168.0.1 dockernet

Second, start a Docker container using NVidia Docker driver, replace your absolute path to cloned project with {absolute path to face_service}: i.e.,'/home/face_service/:/workplace/

NV_GPU=3 nvidia-docker run -d --name=face_dtm -v {absolute path to face_service}:/workspace/ --net dockernet --entrypoint="tail" face:latest -f /dev/null

Finally, you can access to the container:

docker exec -it face_dtm /bin/bash

Some demo:

1. Create office3 folder: mkdir data/videos/office3

2. Decode and extract frames from video: ffmpeg -i {path_to_test_video} data/videos/office3/%05d.jpg -hide_banner . Decode with specific frames ffmpeg -i in.mp4 -vf select='between(n\,55111\,55800)' -vsync 0 -start_number 55111 data/videos/office3/%05d.jpg

3. Run the test python3 main.py

Running the tests

(Upcoming...)

Deployment

(Upcoming...)

Built With

• Tensorflow
• OpenCV
• imageio

Contributing

(Upcoming...)

Versioning

(Upcoming...)

Authors

• Nguyen Thanh Tan - NguyenThanhTan
• Vong Chi Tai - vchitai

License

(Upcoming...)

Acknowledgments

(Upcoming...)