试题详情

资料：Directions : There is 1 passage in this part. The passage is followed by some questions or unfinished statements. For each of them there are four choices , You should decide on the best choice. Your face is the future of smartphone security. Apple made that clear last week when it unveiled the pricey iPhone X. Apple claims to have conquered many of the challenges that have prevented the widespread use offacial biometrics. But a number of computer-vision researchers say they are skeptical that a smartphone-based system like FaceID can account for things like variable lighting conditions or subtle changes in a person"s appearance to create a secure-yet-practical way to unlock a phone a dozen or more times a day. Apple’s new technology does sound promising. The company says FaceID creates a “precise depth map” of one’s visage by projecting more than 30,000 infrared dots against a person’s face, then using the phone’s infrared TrueDepth camera and high-power microchip to collect and analyze the results. Users are also asked to turn their head as they scan so the phone"s machine-learning algorithm can measure the face from several angles and create a more detailed 3-D map of their features. Once the map is created and stored, the iPhone X uses infrared light to help FaceID scan a person’s face even in the dark. Meanwhile, machine-learning algorithms running on the phonekeep track of changes in a person"s appearance—including glasses, facial hair and hats—so the smartphone’s accuracy improves over time. Despite advances in facial recognition in recent years, it remains unclear whether FaceID will work in a variety of conditions while also keeping the iPhone X secure. Hackers, for example, quickly found a way to bypass the Samsung Galaxy S8"s facial-recognition scanner when it was introduced in March: They tricked the device by simply showing it a photo of the user. FaceID’s use of 3-D facial maps could address that problem. But historically it has been a big challenge for such a system to recognize faces under different lighting conditions and from a variety of angles. Apple showed off FaceID last week under relatively controlled conditions, says Arun Ross, a professor of computer science and engineering at Michigan State University. "Clearly the demo was very interesting," he says. "But at the same time some extraordinary claims were made." Apple" s Schiller said, for example, that the chance a random person" s face could unlock someone else" s iPhone X was one in a million——much more secure than TouchI D. Ross says, however, that it is not clear how often FaceID fails to recognize its owner. When contacted, Apple declined to elaborate on Schiller" s comments. “Like all biometrics, FaceID will have a problem with revocation,” says Vitaly Shmatikov, a computer science professor at Cornell Tech. “If a password is compromised, it can be changed—but a face cannot be changed.” Apple touts its ability to secure data on its iPhones, which do not share biometric information with the company’s servers. Still, Ross says, hackers always seem to find a way around even the tightest security. Why does the author mention Vitaly Shmatikov saying "a face cannot be changed" ? A.To infer that a person" s face is highly unlikely to change to another shape.