High-definition glasses-free 3D single-chip best solution

In recent years, with the continuous progress of flat panel display technology and related materials, as well as Hollywood's application and promotion of 3D display technology, 3D display technology is getting closer and closer to us. This pursuit of 3D display is in line with the development law of display technology. In the display field, we have experienced a technological leap from black and white to color, from color to high definition, and now the time has come to leap from 2D display to 3D display.

At present, there are two types of 3D display technologies that we are more exposed to: glasses-type 3D display; 3D display without glasses (i.e. naked eye 3D display). In addition, there are holographic displays, but holographic technology is still difficult to promote. Glasses-type 3D display, as the name suggests, requires users to wear special 3D glasses to form a 3D effect by allowing the eyes to observe images with parallax separately. Glasses-type 3D display is a technical model commonly used in cinemas and 3D TVs. Glasses-type 3D display has a low technical threshold and is simple to implement, but it has disadvantages such as changing the user's viewing method, not conforming to usage habits, reduced display brightness, and is particularly unsuitable for application on mobile devices.

Naked-eye 3D display refers to placing specially designed parallax barriers or lens arrays on the display panel through parallax barriers or lens array technology, and through their light splitting, the images with parallax enter different eyes separately to form a 3D effect. Naked-eye 3D display technology is complex. In addition to hardware, it also requires special image processing algorithms to form a 3D effect. This solution has been recognized by many notebook manufacturers and will be launched in the second half of this year.

Realization of naked eye 3D

The naked-eye 3D display technology adopted by SuperD is to place a uniquely designed birefringent cylindrical lens array in front of the display panel and use specially developed image processing software to create a 3D effect. In addition, SuperD has also specially developed a chip for completing 3D image processing, making 3D image processing independent of existing hardware computing resources. While being able to obtain a good 3D display effect, the utilization rate of existing hardware resources is still guaranteed. Figure 1 is a schematic diagram of naked-eye 3D display. The actual technology is far more complicated than shown in the figure. Figure 1 describes the basic principle of naked-eye 3D display.

Lens-type naked-eye 3D display technology does not reduce display brightness and does not need to change users' viewing habits, making it particularly suitable for implementation on mobile devices. SuperD's naked-eye 3D display technology requires improvements to the existing LCD manufacturing process, the synthesis of new materials, and the use of precision micro-machining technology to make optical molds and devices. It is also necessary to develop specialized software and graphics algorithms to optimize 3D effects and meet users' needs for film sources and applications. In order to balance the 3D graphics and image processing load and retain the development space for completely independent 3D display graphics and image processing in the future, a specialized chip needs to be designed.

Implementation of 2D/3D point-by-point technology

SuperD's 2D/3D integration technology is also an important attribute that distinguishes it from glasses-type 3D. Because glasses-type 3D cannot distinguish between the 3D area and 2D area on the screen, it can only use the same processing method. This causes the 2D display effect to be interfered with by 3D glasses, becoming less natural than before, thus affecting the visual experience.

In order to ensure that 3D display does not change the habits and application effects of 2D display, SuperD has developed a technology called pixel-by-pixel 2D/3D switching (referred to as 2D/3D integration technology). Its feature is that it allows 2D and 3D content to be displayed on the same display panel at the same time. For example, this technology can be used to display online on-demand videos, and the video window can watch 3D movies, while other areas are still 2D displays, including 2D pictures, text, etc. are not affected. Figure 2 describes the effect of 2D/3D integration technology.

The 2D/3D integration technology developed by SuperD includes: 1. A specially designed birefringent optical device that can switch the lens state under a certain driving voltage; 2. A driving circuit specifically designed for this optical device. The design of the driving electrode fully considers the optical interference effect, so as to achieve no impact on the picture effect; 3. Pixel-by-pixel 2D/3D driving for this optical device to achieve the switching, moving, zooming and other functions of the 3D area.

The area marked as "3D window" in Figure 1 shows the 3D effect, and the area outside the 3D window is the 2D text area. The largest application area of ​​2D/3D integration technology is the Internet. Through this technology, people can continue to use 2D functions while enjoying 3D content on the Internet. Because the Internet has penetrated into all areas of life and production, through the Internet we can mobilize everyone to use a large amount of resources to produce, disseminate and share 3D content and 3D applications, which in turn will greatly promote the development of 3D display technology. Therefore, the support of 2D/3D integration technology for the Internet is a very virtuous cycle, which not only ensures the compatibility of 3D display and 2D display under existing technical conditions, but also promotes the continuous development of 3D display.

Head tracking technology

Due to the limitation of the principle of freestyle 3D display technology, a series of viewing areas will be formed within a certain space. In these areas, the audience can freely watch the 3D effect, but if they go beyond these areas, they cannot see the normal 3D effect. There is no unified definition for this phenomenon. SuperD calls this state that the 3D effect can only be viewed normally in certain areas but not outside these areas "shear phenomenon". Figure 3 describes this phenomenon.

In Figure 3, the 3D effect can be seen normally at position A, but the 3D effect seen at position B is wrong. Because the images seen by the two eyes are exchanged, this creates a 3D layering effect that is completely opposite to the layering that should be shown. The human eye cannot automatically adjust to this opposite effect, which causes severe dizziness. The 3D effect seen in the shear area is called "reverse vision". In this case, the audience will find that objects that should be far away are closer to them, and objects that should be near are farther away from them, which is in great contrast to our usual physiological habits.

To solve the problem of limited viewing area, SuperD has developed a 3D display device that integrates head tracking technology into the original 3D display solution. Figure 4 is a schematic diagram of head tracking technology.

SuperD uses tracking technology to obtain the three-dimensional position of the audience, and then adjusts the pixel arrangement of the 3D image according to the position to generate a new 3D image that conforms to the position, thus solving the problem of limited viewing area. The current tracking technology can only be used for a single audience. Head tracking technology itself can track multiple users, but due to the limitations of the refresh rate and resolution of the display panel, it is not suitable for multiple users to use at the same time. This is also the reason why the technology was first applied to laptops and other mobile devices. SuperD's head tracking technology requires very high tracking accuracy, accurate to the centimeter level, and is completed with a camera. This is something that many current head or face tracking technologies cannot achieve.

In addition, in addition to solving the problem of limited viewing area for a single user, head tracking technology can also be used to handle many other applications, such as interaction with display devices, increasing the playability of games in game control, and using eyes to control the mouse to interact with application software, etc. Because the head tracking technology used by SuperD can use a camera to obtain spatial position, it can also be used in the field of measurement.

3D image processing chip

SuperD's 3D image processing chip provides a single-chip integrated solution that can not only receive, process and send display image data streams, but also adjust and control the optical design of the screen according to the different image contents. In the image data path, the chip automatically detects the image format, such as the resolution size, basic parameters of the display image frame, etc. After these parameters are processed by internal calculations, they will be packaged and synchronized again by the image data stream sending module and sent to the display unit of the screen under the premise of ensuring that the data stream bandwidth/throughput rate is not changed, thereby achieving seamless access and processing. At the same time, the synchronization control logic will also control the driving circuit of the screen's birefringent optical device according to the image 2D/3D area in a pre-set working mode, so that it can turn on the refraction effect of the lens under the condition of meeting a specific voltage, presenting a 3D display and a traditional 2D display area. This control of different display areas and image processing must not only ensure control synchronization, but also must take into account the different delay properties of 2D and 3D images to ensure the instantaneous continuity of the image and the matching of bandwidth.

Currently, SuperD's 3D chip already supports full HD resolution and has built-in support for different types of 3D content formats, such as left-right format, top-bottom interlaced format, etc. For upper-level software, this multi-format support can be completed by simply configuring the 3D chip using the relevant configuration interface. Therefore, the realization of this dynamic configuration allows upper-level software to directly operate hardware resources and expand the scope of application.

In image processing, SuperD's 3D chip adopts a unique design concept that uses line cache instead of the frame cache method in traditional image processing chips. This alone can save several times the system power consumption and reduce the cost of the solution.

Implementation of motion parallax technology

To achieve stereoscopic vision, in addition to collecting images with parallax through the two eyes for the brain to synthesize stereoscopic vision, there is another important monocular stereo factor, which is the motion parallax effect. Motion parallax is the difference in the direction and speed of movement of the surrounding objects seen by the viewer when the viewer moves. Figure 5 is a schematic diagram describing motion parallax.

From Figure 5, we can see that when the viewer's position changes, different sides of the object should be seen, and the relative displacement of objects at different distances from the viewer is also different. The application of integrated motion parallax phenomenon can significantly improve the user's sense of immersion. At present, some applications under 2D display have also begun to consider motion parallax phenomenon, so as to create a false three-dimensional feeling to increase the attractiveness of the application.

SuperD will increase the authenticity and immersion of 3D display by combining motion parallax technology with 3D display technology. Therefore, combined with head tracking technology, we proposed a 3D display technology with motion parallax. At present, this technology is still mainly used in game applications. Because the current game development method itself contains three-dimensional description information of the scene, what we need to do is to extract this information and combine it with the audience position obtained by tracking technology to correct the matrix information of the game scene in real time, so as to produce different rendering effects corresponding to different viewpoints, so that users can form a sense of motion parallax, thereby improving the immersion of 3D display technology. Figure 6 is a schematic diagram of the combination of motion parallax and 3D display.

Note the display changes of the two cubes in Figure 6. Here, the 3D display device not only provides a 3D image corresponding to the viewer's position that is newly generated according to the viewer's position, but also processes the 3D image according to the motion parallax requirement.

Z-axis optimization

How to make the simulated 3D world as close to the real 3D world as possible? SuperD proposed a concept called Z-axis optimization. Under the conditions of 3D display technology, due to the introduction of a new dimension (Z-axis), the traditional graphics and image processing technology based on 2D display needs to be improved. Z-axis optimization proposes different graphics and image processing theories and technologies, forming a processing method suitable for 3D display.

First, the Z axis is described using the parallax information implicit in the 3D image. Based on the parallax information, combined with information such as the display panel size, optical design parameters, and the type of scene to be presented, it is determined whether the parallax information needs to be adjusted and how to adjust it. A new 3D image is generated based on the adjusted parallax information, and the 3D effect formed by the new 3D image will be more suitable for display under current conditions. This is the dynamic parallax adjustment technology. Figure 7 describes the principle of this technology.

In addition, based on the parallax information, we can also perform other optimizations, such as image filtering based on depth information and graphics rendering based on depth information. The ultimate goal is to use the concept of Z-axis optimization to optimize the 3D imaging effect and make 3D display technology more adaptable to people's physiological habits.

Conclusion

The naked-eye 3D overall solution is a comprehensive solution. SuperD has conducted in-depth research and development in optics and material research and development, process research and development, chip and electronic engineering, image engineering, 3D vision and 3D applications, etc. This solution will be widely used in various products such as laptops, the Internet, and 3D movies and TV.