Face-Detection-Face-Recognition

First Use Case : Access Control / ID Affectation / Distance Estimation

  • AUTO ENROLLMENT. (DEFAULT : Enabled)

    • Enabled : The user is automatically added to the Database , if he is within the detection threshold.
    • Disabled : The user is labeled as UNKNOWN whithin the detection threshold , Pressing S , adds the user to the Database.
  • ID AFFECTATION. (DEFAULT METHOD : Counter)

    • Counter : Affect a an ID to a user based on a counter.
    • UUID_Generator : Affect an ID to a user based on uuid.
  • DISTANCE THRESHOLD. (DEFAULT VALUE : 50)

    • 50 : Distance at which detection process starts. the unit is CM.
  • FACE DETECTION MODEL. (DEFAULT VALUE : hog)

    • hog : Better suited for normal machines. (FAST)
    • cnn : Better preecision , but requires heavy computations. (USE ONLY WHITH A GPU)
  • NUMBER OF JIITERS. (DEFAULT VALUE : 1)

    • 1 : Default value , how many times the each detected face will be distorted , and manipulated , HIGHER NUMBER == LOW FPS
  • DISTANCE ESTIMATION METHOD. (DEFAULT VALUE : Algebra)

    • Algebra : Uses geometric formulas about eyes positions , to estimate the distance. (Better Estimation)
    • Estimation : Uses special formula to estimate distance.
    • Perimeter : Uses the opencv box perimeter to estimate distance.

IMPORTANT

DummyFuncUnknown :

  • Execute One time , when an UNKNOWN person gets within the detection distance , he wil automatically be enrolled and affected an ID .

  • This function triggers One time per user , after being added to the DB , and return its affected ID.

    • Message Shown on the Console :

       [+] ==> UNKNOWN USER ADDED TO BD. 
       [+] AFFECTING ID : {} TO THE USER
      

DummyFuncEnrolled :

  • Execute whenever a known person gets within detection distance.

  • This function triggers every time a known user is within detection distance , and returns its ID.

    • Message Shown on the Console :

      [+] KNOWN USER.
      [+] AFFECTED ID IS : {}