Super large base+telephoto big upgrade, vivo couldn't resist releasing X90 series camera

According to the previous release rhythm of vivo X series, the release of the new generation X90 series should not be far away. On October 24, Vivo held a video strategy conference. Interestingly, the conference did not explicitly mention any information about the X90, but used the "next generation" to refer to this series of new machines crazily.

To refine the image of the "next generation", the new Vivo machine has these upgrades——

1. CMOS with a larger base, and the photosensitivity is increased by 77%

2. Sensor signal-to-noise ratio increased by 20%, and color restoration increased by 15%

3. The analytical power of the telephoto lens is increased by 64%

4. The new generation of self-developed image chip is upgraded from ISP to AI-ISP

Of course, there is also some blessing of algorithm technology——

1. Color reduction engine, ultra clear image quality engine, computing power acceleration engine

2. Environmental Understanding Technology

3. Sky Night View System

4. Super touching image system

What is more interesting is that many of the samples displayed at this press conference only have the "vivo ZEISS" watermark, without the model logo. In other words... you know.

01   Hardware upgrade, CMOS is expected to be 1 inch?

It has been exposed online before, vivo X90 The series will be equipped with 1-inch oversize CMOS. Vivo made it clear in this press conference that the sensitivity of the next generation of main camera CMOS is 77% higher than that of GNV sensor. You should know that GNV is already a 1/1.3 inch large size sensor, and its sensitivity has been improved so much that it should not be possible to increase its size too small. It is likely to be a 1 inch large bottom. At present, there are rumors that CMOS may be Sony IMX989, but there is no clear evidence.

In addition to the base, vivo also released a "VCS bionic spectrum technology" that can better control image noise performance and color restoration 。 Here we quote the official technical data of vivo——

"When light enters the photosensitive element, it will pass through the color filter first. The color filter spectrum of ordinary image sensors is very different from that of the human eye, which requires complex processing to restore the information perceived by the human eye. This processing process will lead to poor image quality. VCS technology is to improve the color filter so that the original information received by the sensor is constantly close to the human eye, so as to achieve better noise performance and color restoration. The signal-to-noise ratio is increased by 20%, and color restoration is increased by 15%. "

It is worth noting that the values of these two improvements are compared vivo X80 The IMX866 on board may mean that the VCS bionic spectrum technology is applied to the X90 ordinary version.

The upgrade of telephoto is also a highlight of the next generation model. After all, the analytical power can be improved by more than 60%. It is worth noting that there are several keywords—— "Optical supersection algorithm, telephoto lens matching with the next generation product, focal section above 5X" 。

What is optical supersection algorithm? Let's continue to quote the official technical data of vivo——

"Because mobile phone Volume limitation: the original image of the periscope camera will lose 50% of the information after passing through the lens. The "optical supersection algorithm" developed by vivo can recover about 35% of the lost information by modeling and analyzing the defects of the lens. In addition, vivo has coupled Denoise, Demosaic, Debrur and other modules. After more than 100 million scene training, it has increased the depth computing capability of the previous generation algorithm by 10 times, built a new ultra clear image quality engine, and upgraded image quality processing from a single point solution to a systematic AI solution. "

The remaining two keywords are easy to understand, which may mean that the X90 will be equipped with an upgraded telephoto lens, which has at least 5 times the optical zoom capacity.

The last significant upgrade is the next generation of self-developed image chips. We are familiar with the V1 chip developed by vivo. It is an ISP chip, which is used to expand imaging computing power, release the ISP load of the main chip, and serve the needs of photography and video recording. It can also be used as a unique chip. The next generation self-developed chip vivo has not given a clear name, but it is clear that Upgrade from ISP to AI-ISP 。

We still quote a section of vivo's official data to explain AI-ISP——

"Traditional ISPs can handle a large amount of data pipelining with extremely low latency, but can only solve known and specific problems. AI is good at dealing with complex and unknown problems, but the latency is large.

AI-ISP combines the advantages of both, which is equivalent to adding a new AI brain to the traditional ISP chip. The distribution of neurons in the AI brain is network structure, which is more suitable for massive divergent information processing. Compared with the traditional architecture, AI-ISP can directly integrate AI computing into ISP pipeline through direct hardware connection, complete seamless data buffering and processing, and greatly improve processing capacity and processing energy efficiency.

Based on the architecture innovation of AI-ISP, vivo has achieved three breakthroughs in the next generation of self-developed chips: breakthrough in on-chip memory unit, breakthrough in AI computing unit, and breakthrough in image processing unit. "

Generally speaking, AI-ISP Faster operations with lower latency (Data throughput rate reaches 1.3 trillion bits) Higher energy efficiency ratio (The peak energy efficiency ratio reaches 16.3 trillion operations per watt). A better shooting effect can be achieved by combining new algorithms. For example, the AI-NR noise reduction algorithm has improved the night scene noise reduction effect by 20%, and the HDR image tone fusion algorithm has increased the dynamic range by 4EV compared with the previous theory.

02   Large collection of algorithm technology, full upgrade of portrait landscape

In addition to the hardware upgrade, the algorithm technology of the next generation products has also been systematically upgraded. First, there are three "engines". We quote a paragraph of official technical data of vivo——

"[Color Restoration Engine]

The main supporting technologies are intelligent AWB and intelligent white plus black subtraction, which can improve the accuracy of white balance and the exposure accuracy of large areas of "white" and "black" scenes, and optimize the color performance and tone perception of vivo to the extreme.

[Ultra clear image quality engine]

The typical technology of the ultra clear image quality engine is the optical hyper division algorithm. Through modeling and analyzing the defects of the lens, about 35% of the lost information was recovered. The optical super division algorithm is the starting point for achieving ultra clear image quality. The engine also has Denoise, Demosaic, Debrur and other modules, which are coupled to train, forming a systematic AI image quality scheme - ultra clear image quality engine. It can realize multi frame image quality reconstruction to achieve better image quality and wider dynamic range.

[Computational Force Acceleration Engine]

Computational force acceleration engine is a set of software hardware combination acceleration scheme. Vivo has designed and developed CV heterogeneous acceleration engine and vDNN deep learning acceleration engine based on self-developed chip and platform SOC. It is an acceleration engine for traditional CV algorithm and AI algorithm respectively, which can support heterogeneous acceleration of multi-core combination and achieve multi information collaboration and high-speed processing. "

The latter two are related to the content mentioned above, and we will not repeat them here. In the past, when we shot scenes with more white, we often felt that the straight photos were gray, while the scenes with more black, on the contrary, the pictures were white and could not be restored to the naked eye. The color refining restoration engine can perform intelligent white plus black subtraction, so that the exposure accuracy is higher and the image perception is more transparent.

Literally, there is a more intelligent technology, namely, environmental understanding technology. This technology allows the device to recognize the environment scene, and more accurately shoot and judge the auto focus, auto exposure, and auto white balance based on the information extracted from the environment.

The sky night view system can tell what it is used for by its name, which is the official explanation of vivo——

"The firmament night scene system has undergone a comprehensive AI upgrade, and three new models have been trained based on millions of night scene photos: super sensitivity model, full segmentation semantic model, and color enhancement tone model. Super sensitivity model: the sensitivity has been improved by 100%, and ISO can support up to 102400; The full segmentation semantic model and the color enhancement tone model are combined to achieve adaptive tone and color adjustment capabilities. "

Of course, there is no shortage of Vivo's great skill in image shooting - portrait shooting. In this regard, vivo has launched a super moving portrait system, which includes three major links: portrait understanding, portrait beautification, and portrait atmosphere——

"The first step is to understand the portrait -- to understand the portrait information like a photographer

Vivo's "Portrait Understanding Technology" will, like professional graphic artists, perform partition detection, extract the semantic focus and contour details of the face, and establish the key face coordinates of up to 103 feature points. The point accuracy error is less than 2 pixels. At the same time, the key points of skin, gestures and limbs in the picture will be detected, and the accuracy of the subject segmentation IOU has reached 96.15%, providing fine processing capability for static portrait remodeling and dynamic portrait capturing.

The second step is to beautify the portrait - beautify the details of the micron level portrait

Vivo uses millions of materials of different ages, genders and scenes, and through repeated machine learning and growth, introduces Vivo's newly upgraded micron level skin rejuvenation technology, which makes the skin look more natural and healthy, and also better retains the three-dimensional sense of the five senses.

Step 3: Portrait atmosphere - create a unique atmosphere

Vivo thinks more about portraits than about people. The relationship between people and scenery is another important dimension of portraits. Vivo "Portrait atmosphere integration" technology, based on the semantic understanding of the portrait, carries out the integration of people and scenes at the levels of tone, color, brightness, virtualization, etc., to achieve the harmonious and natural effect of people and scenes. "

Vivo mobile phone The portrait image configuration of Vivo has always been very luxurious. The press conference mentioned that Vivo has professional portrait lens, super moving image system, front soft light selfie, ZEISS lens bag and other advantageous configurations. Whether all the vivo X90 series are equipped is unknown, but we can look forward to it.

Write at the end

Of course, the above is just the important content I think, including SuperRAW, HDR shadow and tone fusion algorithm, professional LOG mode, etc. The "three comparisons" that Vivo's image technology matrix pursues undoubtedly demonstrate confidence: the comparison of professional image equipment, professional photography team, and professional post capability. It can be said that the video strategy conference of Vivo this time showed its muscles, which also makes the "next generation" model, the Vivo X90 series, more and more attractive.

By the way, Vivo also has a general image strategy, so you can extract keywords by yourself——

"Centering on the" three shoulder to shoulder "strategy, adhere to the two strategic fulcrums of joint innovation and independent innovation, create a competitive advantage in the increasingly homogeneous image arena, strive to build technical barriers on the mobile image track, and maintain a continuous lead."

This article is an original article. If it is reproduced, please indicate the source: Super large base+telephoto big upgrade, vivo couldn't resist releasing X90 series camera https://mobile.zol.com.cn/805/8051588.html