For today's people, there are too many secrets and information hidden in smart phones: from some of the most commonly used contact information around us to various financial information. So safety is particularly important. Everyone needs a unique password to protect their information. What is truly "unique" is ourselves. Therefore, a higher level of security protection has appeared on mobile phones: biometrics.
It has been five years since 2065 438+03 iPhone 5S successfully popularized fingerprint identification to mobile phones. In the past five years, the form of mobile phones and the functions that need to be carried are gradually changing. Under the background of full screen, 3D structured light is gradually connected to fingerprint recognition. Compared with fingerprint identification, 3D structured light can bring a more insensible identification process and a safer checkpoint. More importantly, 3D structured light not only has biometrics, but also can establish image information.
Therefore, at this stage, even if Face ID will have a controversial impact on product design, Apple still insists on doing so.
We know that the traditional camera captures information based on 2D plane, so biometric identification based on front camera is actually very easy to be deceived by false information. Therefore, when the image changes from 2D to 3D, it has security features. Therefore, both three-dimensional structured light and TOF of living body are essentially the acquisition of image depth information.
However, 3D structured light is still different from TOF, although the realization purpose is the same and the realization process is different.
▲ Image from Yole Developpement
The three-dimensional structured light used by Face ID is realized by speckle structured light coding. Within a certain distance, the dot matrix projector will emit 30,000 randomly and discretely distributed infrared light points, covering a space area. When the user's face enters this workspace, the light spot will appear different displacement and distortion, and the mobile phone will record the user's face in the space and draw an accurate and detailed depth information map. When the user makes each recognition, the new "depth map" is compared with the standard "depth map" recorded in the mobile phone in advance, and the mobile phone naturally knows whether it is unlocked by the real user.
TOF used in the body is the abbreviation of flight time. The infrared emitter continuously emits light pulses, which are reflected after encountering people or objects. When the sensor receives the reflected light, it calculates the time difference between infrared light emission and reflection, thus forming stereoscopic vision. Subsequently, the CPU of the mobile phone analyzes the light time difference data to achieve the purpose of establishing a face model. On the key data, TOF needs to collect the effective depth information points of 300,000 faces, and the working distance can be up to 3 meters.
It can be seen that the key of structured light lies in "light", while the key of TOF lies in "distance".
Therefore, in recognition, 3D structured light needs to create a "space", so it needs at least 25mm recognition baseline, while TOF needs almost no distance. The maximum working distance of TOF is 3 meters, while the optimal working distance of structured light is 25-50 cm. In addition, both 3D structured light and TOF are essentially to obtain the depth information of an object in space, so the number of points recorded by TOF in the body is 300,000, which is really more helpful for the edge details of the constructed stereoscopic image. Therefore, from the perspective of technology to function transformation, TOF is undoubtedly more scalable than Face ID.
Moreover, in ID design, TOF camera is smaller than structured light. In the product ID design, the fuselage does not need to pay longer bangs and more internal space for stereoscopic vision. Generally speaking, a set of t of needs an infrared transmitter, receiver and camera, not as many things as face ID.
However, TOF is not without its drawbacks. As a 3D image, the resolution of TOF in two dimensions (namely X axis and Y axis) is not as high as that of 3D structured light. But there is no doubt that both of them have reached the financial level of security, which is enough for ordinary users. Vivo has also announced a payment cooperation with WeChat.
PS: Let me add one more sentence here: At the vivo TOF 3D super-induction conference, it was demonstrated that 3D modeling and even 3D printing can be directly carried out with mobile phones. So many netizens asked what is the difference with Sony's own 3D master. Sorry, we really didn't find the TOF-related information of Sony 3D Master, so it is speculated that Sony 3D Master was completed by XZ 1 laser focusing (laser ranging)+camera+a set of Niu X algorithm. After all, Sony, which acquired softkinetic, is already first-class in 3D imaging and TOF technology.
The future of flight time
▲ TOF prototype can be directly modeled.
Of course, technology on paper is a tree without roots. Combined with the known characteristics of TOF, such as large depth information and long working distance, it can be placed in various use scenarios of life. For example, taking 3D photos through TOF 3D extrasensory technology to build a more accurate human model. On the other hand, the seller preset the information of clothes in advance, so that users can try on clothes directly online to avoid buying clothes with inappropriate sizes online.
On the other hand, using t of camera to model real objects, and then integrating the model with other virtual or real scenes to create more sci-fi MR games is also a major direction of TOF's future development. For example, model you and your friends, and then put them on a virtual runway or other interesting scenes.
The boundary between fuzzy reality and virtual reality is the force point of t of in the future.