[API][yes]
[Search][yes]
[UI][yes]
Face verification engines analyze human faces in media assets and score them as to similarity with the face(s) of a specified user identity.
Training and libraries
Training for the face verification engine is done by using the enroll mode of the engine, specified when calling the engine. The hashed faceprint of the trained identity is stored in the library, with the hash key stored in an additional database.
[Note] The verify mode of the engine retrieves the hashed faceprint from the library corresponding to the specified username, decrypts the faceprint using the hash key, and compares the decrypted faceprint to the faceprint extracted from the input image.
Engine input
The face verification engine uses segment processing for its processing mode. It accepts as input a custom binary file containing the following in the respective order:
- 8 bytes containing the number of bytes of a byte-encrypted json string
- A byte-encrypted JSON string
- 8 bytes containing the number of bytes of a binary image file
- Binary image file
An example of the byte-encrypted JSON string is as follows:
{
"mode": "verify",
"username": "jsmith@veritone.com",
"libraryId": "13e6f4a3-0d5c-4e11-9a30-913e981cb9ad",
"dbUser": "postgres",
"dbHost": "127.0.0.1",
"dbDatabase": "postgresdb",
"dbSchema": "public",
"dbPort": 5432
}
Engine output
The face verification engine output should be stored as an object in the AION. The type of the object is verification. Each face maps back to a specified user identity which corresponds to an entity in a library; hence the object includes the entityId along with the libraryId. The similarity score of the face to the face(s) for the entity is the confidence. The mode specifies whether the engine is run in enroll or verify mode.
Example
Here is an example of the simplest type of face verification engine output:
{
"schemaId": "https://docs.veritone.com/schemas/vtn-standard/master.json",
"validationContracts": [
"verification"
],
"object": [{
"type": "face-verification",
"mode": "verify",
"entityId": "11a14999-0531-4d3e-9a44-68cdd4f93659",
"libraryId": "13e6f4a3-0d5c-4e11-9a30-913e981cb9ad",
"confidence": 0.80
}]
}
