LaTalamus – Contactless Attendance
Artificial-Intelligence based face recognition attendance system. This product involves building an attendance system that utilizes facial recognition to mark the presence of IN and OUT time. The aim of this product is to automate the traditional attendance system and the important feature of this product is Contactless that “Attendance will be marked without any contact with any card or machine”. It also enables an organization to maintain its records like Time status (IN and OUT).
Functions
Capture Faces
Get input from the user (manually enter a record in the database)
Bulk insertion from another source.
Train faces according to their unique Id's and saves their 128-D features in the database.
Capture Faces
In order for face recognition functionality to work the admin user needs to enter an employee’s face, which can be achieved by capturing the image, followed by a comparison of extracted image data with existing image data. The program will generate a GUI window and by clicking on the Input/LoadData button user will enter the employee id and employee name to display the real-time image obtained by the camera. Then capture the face according to keyboard input by pressing the designated buttons i.e.
‘N’ - Enter a new face, create a new folder, Record insertion in Database
‘S’ - Starts to capture the face.
‘Q’ - Exit window


Extract Features / TrainDataSet

Face Recognition/Attendance (Video Stream)

Android Version
How it works
User logs in by entering their credentials
Press the evaluate button
Camera turns on
If found any face capture the image.
Send the image to the evaluation engine through API
In return show the person ID on the screen.
After a delay of 3-5 sec, the streaming mode is turned on again, ready to capture the face object.
If any unknown face is detected an error message is displayed on the screen.

Tools and Technology





