To load captured or reference faces from the database, do the following:
Click the Run button to load the data to the folder (6). The total number of faces loaded from the database will be displayed.
The data will be loaded to folders corresponding to the names of video cameras by which the faces were captured.
Two files will be loaded for each captured face – the corresponding vector in .bin format and the image of the detected face in .jpg format. The name of each file has the following structure: [date and time] - [ID] - [recognition module].
The faces added to the Face Recognition Server database are loaded to the Path: folder. For each .jpg file with an image of a captured face, a .json file with the same name will be placed in the same folder. The .json file contains the user's full name, department, comment and other data that are taken into account while loading images from the folder to the face database (see Face Intellect Operator's Guide, the Upload an image to the reference face database section).
If there are forbidden characters (\/:*?"<>|) in the name of the reference face file added to the Face Recognition Server database, then these characters will be replaced with _ character. |