Apple wins patent for automatic retrying of a failed facial recognition authentication process
The United States Patent and Trademark Office granted a patent for Apple on May 28th, 2019 that relates to retrying of a facial recognition authentication process after the facial recognition authentication process fails at an attempt to authenticate a user.
Apple notes in the patent that biometric authentication processes are being used more frequently to allow users to more readily access their devices without the need for passcode or password authentication.
One example of a biometric authentication process is fingerprint authentication using a fingerprint sensor. Facial recognition is another biometric process that may be used for authentication of an authorized user of a device.
Facial recognition processes are generally used to identify individuals in an image and/or compare individuals in images to a database of individuals to match the faces of individuals.
Apple further notes, for authentication using facial recognition, the facial recognition system may sometimes encounter problems in authenticating an authorized user when images captured during the authentication process are captured under non-ideal conditions.
For example, the user’s face being too far/too close to the camera, the user’s face having some occlusion in the captured image, and/or the user’s attention or pose in the images being less than ideal may prevent matching (e.g., authentication) of the authorized user in the captured images with the authorized user enrolled on the device.
If the user repeatedly fails to be authorized using the facial recognition authentication process, the user may become frustrated with the experience and look for other avenues of authentication and/or search for another device to use instead of the current device.
Automatic retrying of a failed facial recognition authentication process
In certain embodiments of the Apple’s invention, in the event a facial recognition authentication process fails to authenticate a user, the facial recognition authentication process may re-initiate and retry to authenticate the user using newly captured images.
The re-initiation of the facial recognition authentication process may be automatic without input from the user (e.g., the user does not have to provide additional input to re-initiate the process). Automatic re-initiation of the facial recognition authentication process may provide a more satisfying user experience (if the re-attempted process is successful in authenticating the user).
In some embodiments of the Apple’s invention, one or more criteria for the images are adjusted when the facial recognition authentication process is re-initiated. The criteria for the images may include criteria that are useful in determining that the images can be successfully operated on to authenticate the user.
Examples of criteria for the images include distance between the camera and the user’s face, attention of the user in the images, pose of the user’s face in the images (e.g., pitch, yaw, and roll of the face), and/or occlusion of the user’s face in the images.
Adjusting one or more of the criteria before re-initiating the facial recognition authentication process may improve the chances of success in the subsequent facial recognition authentication process.
Apple patent’s FIG. 4 shown below depicts a flowchart of an embodiment of an image enrolment process for an authorized user of a device.
Process #200 may be used to create an enrolment profile for an authorized user of device.
Process #200 begins with authenticating the user in #202. In #202, the user may be authenticated on device #100 (not shown in the figure) using a non-facial authentication process. For example, the user may be authenticated as an authorized user by entering a passcode, entering a password, or using another user authentication protocol other than facial recognition.
After the user is authenticated in #202, one or more enrollment (e.g., reference or registration) images of the user are captured in #204.
After the multiple enrollment images are captured in #204, selection of enrollment images for further image processing may be made in #206.
After images are selected in #206, features of the user in the selected (template) images may be encoded in #208. Encoding of the selected images may include encoding features (e.g., facial features) of the user to define the features in the images as one or more feature vectors in a feature space. Feature vectors #210 may be the output of the encoding in #208.
Process #200 may include, in #214, storing feature vectors #210 in a memory of device #100 (e.g., a memory protected by SEP #112).
In certain embodiments, feature vectors #210 are stored as static templates #216 (e.g., enrollment templates or reference templates) in a template space of the memory (e.g., template space #220 described below). Static templates #216 may be used for the enrolment profile created by process #200.
Apple’s FIG. 7 shown below depicts a flowchart of an embodiment of a facial recognition authentication process.
Adjusting one or more selected criteria in #310 for re-initiations of process #250 may provide a more satisfying user experience for the user.
The experience may be more satisfying as the re-initiated attempts (e.g., retry attempts) of process #250 to authenticate the user may have a higher chance of success (if the user attempting to unlock device #100 (not shown in the figure) is the authorized user) with the tighter tolerances placed on the selected criteria.
Providing higher chances of success for process #250 may decrease the frequency that the user has to use a secondary authentication process to unlock device #100.
The user may also experience faster unlocking of device #100 with automated retry attempts of process #250.
These factors may increase the satisfaction of the authorized user in the facial recognition authentication process, thus increasing the usability of the facial recognition authentication process for the user.
This patent 10303866, granted to Apple on May 28th 2019 was filed on September 25th 2018.
Let us know your thoughts in the comments section.
Subscribe to our newsletter to stay updated on latest inventions.