Is the era of 3D photography coming？ Three-shot analysis of OPPO R17 Pro TOF

　 　 Before the advent of Pro, the only popular three-camera solution on the market was Leica Three-camera, which Huawei was proud of. Huawei adopted a three-shot scheme of wide angle+telephoto+black and white. This three-shot scheme is the evolution of Huawei after the wide angle+black and white dual-shot scheme. The telephoto lens is added to solve the problem of zoom shooting clarity and is a further optimization of the traditional shooting scheme.

The top lens is the TOF camera.

　 　　When OPPO announced the adoption of the three-camera scheme at the press conference, I thought that OPPO also adopted the three-camera scheme of telephoto plus black and white, and OPPO should also improve the image quality through multi-camera combination, but unexpectedly, OPPO’s three-camera scheme has come out of its own unique way-TOF lens+main lens with variable aperture+depth of field three-lens design, which is unprecedented. The variable aperture has appeared on the previous Samsung flagship, and the depth of field lens has become very popular. What is the TOF lens and what is its function? Let me answer the secret of TOF for you.

　 　　TOF, also known as time of filght, is a sensor that calculates the depth of field of an object by measuring the reflection time difference of infrared light. TOF sensor is not a new technology that has just appeared recently. It has been widely practiced in the fields of logistics, security, medical care, driverless and so on. As early as the MWCS exhibition at the beginning of the year, vivo had already demonstrated a concept to realize facial recognition similar to Apple’s 3D structured light by applying TOF to the front camera. But what I didn’t expect was that OPPO actually put TOF and the rear lens together to form a new three-shot scheme.

　 　 Regardless of the principle, let’s see how TOF modeling of R17 Pro is realized. The first step is to open the 3D modeling function and establish the model plane datum.

Establish model position

　 　 After establishing the model position, we can press the shutter to start scanning, and the scanning effect can be fed back in real time.

Rotate the phone to complete the object shape scanning.

　 　 After scanning, just press the stop key to view the scanned model directly.

The models in the photo album can be freely rotated and viewed.

　 　 What are these magical operating principles? Let me explain TOF’s 3D modeling technology in detail. First of all, we want to talk about the depth of field measurement principle of TOF camera:

　 　 1. The fill light of 1.TOF camera will actively illuminate the subject.Emit a modulated special infrared signal.

TOF emits infrared light when used.

　 　 2. By calculating the phase difference between the emitted light and the reflected light (that is, the time difference), the depth of field and distance data between the camera and the object can be obtained, and then the 3-D model of the object can be obtained after conversion.

Depth of field map of cans (Source: Texas Instruments)

　 　　This principle sounds no different from 3D structured light. It also uses infrared reflection time difference to calculate distance/depth of field information and complete 3D modeling. However, because 3D structured light is projected by coding, the projection range is small, and the projection distance is short, so high accuracy can only be achieved at a very close distance. At the same time, when 3D structured light projects a bitmap onto the surface of an object, it will produce serious reflection, so 3D structured light can only be applied to front face unlocking.

3D structured light projection (picture from: android.poppur)

　 　　TOF is the projection of surface light source, which projects infrared light evenly on the object, and then uses TOF camera to capture the time difference of infrared light reflection, and obtains the depth of field information through calculation. This has the advantage that the projection imaging distance is far and the accuracy is not bad. Because the active light source is also used, TOF imaging in dark environment is not affected. Moreover, placing TOF in the rear camera area can make it easier to place the active light source and reduce the squeeze on the mobile phone space. Crucially, TOF technology is relatively mature, and its cost is lower than that of 3D structured light.

Comparison between TOF and 3D structured light (data from: android.poppur)

　 　　Of course, but the rear TOF is obviously not designed for face recognition, but the shortcomings are not enough to offset the advantages of TOF imaging distance. TOF combined with camera will certainly bring more games, such as mobile phone modeling scanning, AR changing clothes, somatosensory games and other applications, which will enter a rapid development path in the future. (You can read this document if you want to know more about TOF technology.)

　 　　TOF is here. The specific gameplay is believed that OPPO is already under development. The variable aperture equipped with R17 Pro is also the first time in the domestic Android camp. What is the difference between this variable aperture lens and the traditional lens? Let’s talk about it in detail.

　 　 This time, OPPO used a variable aperture on R17 Pro, in which the main camera used a 12-megapixel variable aperture design, and the variable aperture automatically switched in the two-speed aperture of f/1.5-f/2.4. In addition to using portrait mode to forcibly start the large aperture of f/1.5, under normal shooting, the aperture size is automatically switched by the system according to the ambient brightness and other factors. Let’s take a look at how the variable aperture works.

Note the variable aperture in the middle.

　 　　When the ambient light is very bright, R17 will automatically reduce the aperture size. When the ambient light is dark, the mobile phone will enter the f/1.5 aperture to increase the light intake. In portrait mode, the camera will force the large aperture of f/1.5 to improve the blur effect. The sub-camera below the main camera is designed with a 20-megapixel f/2.6 aperture, which mainly assists the portrait mode to blur the cutout. The main lens not only has a variable aperture, but also is equipped with OIS optical image stabilization and supports the handheld super night scene mode. Let’s take a look at a set of proofs to see how the hand-held night scene mode works.

Handheld night scene mode is not turned on.

Turn on the handheld night scene mode.

　 　　In the hand-held night scene mode, the brightness of photos has been greatly improved, and the suppression of highlights is also in place. It can be seen that the lights basically did not overflow, the dark details were also preserved, and the architectural details were completely preserved. It is worth mentioning that in the normal photo mode, AI scene intelligent recognition will detect the picture scene, and when the environment is too dark, it will automatically turn on the night scene mode and take a large-scale photo. It is no longer a dream to send a circle of friends.

　 　　This time, OPPO’s adoption of TOF depth-of-field camera on the rear lens can be said to be the first step to take photos in 3D. Of course, what kind of application TOF depth-of-field lens can achieve in the rear area of the mobile phone still needs OPPO to explore slowly.

　 　 At the current level of technology, it is still a little difficult to realize industrial modeling of mobile phones, but after all, this step has been taken. Perhaps in the future, mobile phones, TOF depth-of-field lenses will become standard, somatosensory games with 3D models, and AR dressing needs 3D modeling applications to show their talents. Perhaps the era of mobile phone 3D photography has arrived. Let’s look forward to the future 3D photography era.