Skip to content

Facial recognition where an image of a person reocnoce and validates it through the camera, ie compares what you are seeing in the camera with the image if they are the same comes out the label of the selected name and if not comes out "False".

Notifications You must be signed in to change notification settings

diegoperea20/Face-recognition-photo-to-camera

Repository files navigation

Face recognition photo to camera

Facial recognition where an image of a person reocnoce and validates it through the camera, ie compares what you are seeing in the camera with the image if they are the same comes out the label of the selected name and if not comes out "False".

StepLast

Steps to implement

Note: Have good GPU

  1. Modify root to your root 'dlib-19.22.99-cp310-cp310-win_amd64.whl'

StepLast

  1. Use Dockerfile
  2. Use virtual enviroments and apply requirements.txt
#virtual enviroment with conda 
conda create -n my_enviroment python=3.10.12

pip install -r requirements.txt

if you do not want to use requirements.txt

#install dependecies o libraries
#Same root 
pip install dlib-19.22.99-cp310-cp310-win_amd64.whl

pip install face-recognition
pip install opencv-python

Another way to do it but more time consuming or error with dlib

Steps
  1. Install Visual Studio

StepLast

StepLast

  1. Without virtual enviroments , in local cmd
#install dependecies o libraries
pip install cmake

pip install dlib 

For all packets to be transmitted to a virtual environment such as conda is

conda install -c conda-forge dlib
  1. You can now create a virtual environment and make
pip install face-recognition
pip install opencv-python

Or if is local

pip install face-recognition
pip install opencv-python

About

Facial recognition where an image of a person reocnoce and validates it through the camera, ie compares what you are seeing in the camera with the image if they are the same comes out the label of the selected name and if not comes out "False".

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published