Ghost face detector 2.4.5

Dear users!

While analyzing the face image (face tracking), it has been observed that standard face tracking algorithms sometimes detect faces where there are none. The contours of the face appear for a moment in places of the frame where there is no image of a person's face. Initially, I thought this was an algorithm error, and in many cases, it is. However, during one experiment, I interacted with the ghostly outline of a face that appeared. I asked it to freeze, and to my surprise, the contour stopped disappearing. Then I asked it to open its mouth and nod its head to the left, and it responded accordingly. The likelihood of this being a coincidence is very low.

I decided to share this discovery with you and created a program that includes only the algorithm for tracking a person's face and an additional service. The program offers the following features:

• Face tracking during camera operation.
• Switching the camera from front-facing to selfie mode.
• The ability to record videos with or without the ghostly contour.
• The ability to take photos with or without the ghostly contour.
• View captured videos or photos and search for ghost faces in them. The videos or photos can be taken either by this app or by any other camera app.
• Share your findings with friends by sending them through the app's standard "Share" service.
• Receive videos and photos from friends and analyze them using the app.

In the Pro version, additional features are available:

• Choice of detector type: basic face detector, the second face detector from the Dlib library, or cat detector.
• Adjustment of sensitivity for all types of detectors.
• Setting the type of output for the face contour.

The app does not have any connections to external servers (except for Unity ADS), it does not require logins, and it does not secretly send any information. The main purposes for camera and memory usage are limited to capturing and playing photos and videos for tracking entities such as ghosts or spirits.

The application utilizes Unity3D, openCV, cascade haara, and dlib technologies, which are widely used in popular applications that apply overlay masks on the face.

There are a few limitations:

• The algorithm detects faces that are looking directly at the camera or at an angle of no more than 20°. Ghosts, who are not looking at the camera, will not be detected.
• The algorithm requires considerable CPU resources and may slow down when processing large recorded videos or photos. However, it still analyzes each frame, albeit in slow motion.

Please be patient as ghost faces may not be found immediately.

Wishing you a successful search,

Michael Frankel

P.S. It is possible that this is an error of the algorithm. The program does not claim to be scientifically accurate; it is intended more as a playful experience.

MY BELIEF:
!!! NO WAR!!!