vk_face_recognition

A program for creating a database with human biometrics based on photos. Photos will be automatically downloaded from VKontakte. Subsequently, it will be possible to search for a person using the created database. The program was created based on the face_recognition and vk_api libraries

Installation

Requirements

• Python 3.8+
• Linux, Windows or macOS

Installing on Linux or Mac

pip3 install vk_face_recognition

Installing on Windows

pip install vk_face_recognition

Usage

Initial setup

To get started, you need to add your token_vk received on the site, to the file main.py to a string

self.session = vk_api.VkApi(token='your_token')

Using the Library

from vk_face_recognition import Vk_recognition

vk_recognition = Vk_recognition()
vk_recognition.start_recognition(age_min=25, age_max=26, gender=1, city='Moscow')

When running this code, a file will be created first users_id.txt with the user id according to the specified parameters of age_min, age_max, gender and city. Then the user's photos will be downloaded to the data folder. And then these photos will be used to recognize faces and create face parameters and write them to a separate file in the pickle_files folder

Search by photo

After creating the database you need, you can search through the photo in the following way

from vk_face_recognition import Vk_recognition

search = Vk_recognition()
search.find_people('path to the photo')

A search will be performed on the database you created and, if it matches, it will report this