DeepFaceRecognition

DeepFaceRecognition repository provide an ability to detect faces on images. To use our project follow this instruction.

Installation

$ git clone https://github.com/yur1xfd/DeepFaceRecognition.git && cd DeepFaceRecognition

You can use our project in 2 ways.

Set a TZ:

$ export TZ=Europe/Moscow && ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone

Make sure that make and git is installed in your system:

$ apt-get update && apt-get install -y make git

Install all dependencies:

$ make prereqs

Build docker image using Dockerfile

$ docker build -t deepface .

Run builded image

$ docker compose up -d

Attach to container

$ docker attach deepface-service-1

Put images you want to use for face recognition to input_raw/ folder. To run our DeepFaceRecognition pipeline:

Build executables

$ make build

Run preprocessing stage

$ make preprocessing

Run processing stage

$ make processing

Run postprocessing stage

$ make postprocessing

Now you have the results in the output/ folder.

Also you can run full pipeline with one command:

$ make run

Here an example of running our pipline:

Raw input image
Preprocessed image after make preprocessing (Resize from 1920x1080 to 960x540)
Output image after processing and postprocessing stages

Foundations of Software Engineering for AI course final project
Our team: