Apple has also responded to many questions about Face ID since the release of iPhone X.More specific instructions, which explains that Face ID will be designed using the same protection framework as Touch ID, emphasizing that all information is stored only on the user's phone for comparison purposes and will not be used for other purposes or provided to third-party apps. Regarding Face ID's learning mechanism, Apple also explained that it will record subtle changes as the user continues to use it. Therefore, even if the user gains or loses weight over time, or uses Face ID to unlock in different lighting conditions, the system can continue to "learn" the user's actual appearance.
The Face ID facial recognition mechanism used in the iPhone X has sparked considerable discussion and skepticism since its release. Apple earlier clarified the technology, emphasizing that it utilizes the same privacy protection framework as the Touch ID fingerprint recognition mechanism. This means that all information is stored only on the user's phone and, aside from being used for data comparison, is not used for any other purpose, nor is it stored on Apple servers or elsewhere. Furthermore, it is not shared with third-party apps or organizations.
As previously reported by Apple's senior vice president of software engineering, Craig FederighiExplainTo ensure Face ID operates properly, if the system detects too many faces in front of the camera, or an abnormal recognition situation (for example, if the user's eyes are closed or the user's registered information does not match), the system will switch to requiring a numeric passcode to confirm the actual identity of the person using the phone. Previously, when using Touch ID fingerprint recognition, if the user's finger was sweaty or oily, causing the data comparison to be abnormal, the system would actually require a numeric passcode as a second verification step.
forApple Privacy PolicyFrom the looks of it, Face ID and Touch ID use the same information privacy protection mode, so App developers do not need to make any special changes to the original design. At the same time, Face ID analyzes and compares the user's facial details through a 3D depth-of-field infrared lens module, emphasizing that the overall recognition error rate is only 100 in 1 million. Compared with Touch ID's error rate of 5 in 1, it can more clearly identify the user's identity, and even the subtle differences between twins can be clearly distinguished.
If the eyes are closed, or if the user is looking through a realistic mask or photo, Face ID recognition will basically not work properly. However, if the user becomes fatter or thinner, or if there are any subtle changes in the facial features, such as the lighting in the environment affecting Face ID recognition, the system may switch to requiring the user to enter a numeric password to confirm the identity, and the A11 Bionic processor will memorize and learn it. When a similar situation occurs again, it will use deep learning experience to determine whether the user is the same person.
Therefore, Apple emphasizes that Face ID can still recognize users normally even if the user changes their clothing or body shape, because the system will continue to learn subtle changes in the user's facial features in different situations through comparison, rather than just using the initial Facebook login information as the basis for judgment. At the same time, Apple also emphasizes that all learned information will only be retained in the user's mobile phone and will not be used for other purposes.
According to Apple, when Face ID first logs in to Facebook, it will require the user to enter information about the side of the fingertip, just like Touch ID. It will also require the user to record Facebook information at different angles. If an error occurs in the subsequent judgment, the user can continue to confirm his identity by entering a digital password, allowing the system to record more Facebook recognition information, thereby improving the accuracy of Face ID recognition with the number of times it is used.
However, if the facial features of twins are very similar, or if the facial features of infants and young children are still growing up and there are no obvious differences, Apple also admits that it is still possible to successfully deceive the Face ID system. However, such situations are actually still relatively rare exceptions. For example, it is less likely that ordinary people can use the facial features of infants and young children to unlock the phone, and twin users can still choose different identity recognition methods to ensure personal privacy.
