Epson Discusses Hardware Technologies for 3D Displays
Three-dimensional (3D) TVs are expected to be commercialized in 2010. How much 3D technologies will advance? How will they be used for mobile phones and other devices? We asked these questions to Goro Hamagishi, who is a general manager of Epson Imaging Device Corp and has been engaged in the development of 3D displays for more than 15 years.
What do you think about the current status of 3D displays?
As for displays using special glasses, they are almost ready to be used at home thanks to 3D Hollywood movies. On the other hand, in regard to displays that do not require glasses, some prototypes have just become sophisticated enough to satisfy general consumers in terms of image quality. And there aren't many applications for them.
Thus far, the focus of 3D display development has shifted from displays using glasses to two-viewpoint naked-eye types and to multi-viewpoint types. Though companies have not been able to find killer applications so far, they are now commercializing 3D displays using glasses because of 3D movies.
It is possible to make two-viewpoint 3D video from movies and TV pictures. And video games contain 3D data. It became possible to make 3D video in real time thanks to the improvement of processors. So, we can now tap the 3D display market together with the content industry.
Even 3D displays using glasses reportedly have a problem that an image for one of the right and left eyes can be seen by the other eye (cross talk). And the displays and glasses need to be improved to enhance their comfort and safety.
I think those problems have almost been solved by improving liquid-crystal shutters and polarizing plates. And further improvements can be made by advancing the development of devices. However, even if the performance of 3D displays is improved, problems such as motion sickness might occur due to the low quality of contents.
I am not engaged in the development of displays using glasses. But, in the case of 3D displays, not only displays but contents will influence the safety. So, it is important to evaluate both hardware and contents to enable to view 3D displays safely.
Please explain about the naked eye-type displays?
In the case of the naked-eye types, the number of viewpoints varies from two to 256. But, because the resolution of display does not change, the resolution of the image seen from each viewpoint decreases as the number of viewpoints increases. So, I think it is important to optimize displays for each application.
In the field of ergonomics, one of the issues of 3D displays is to decide whether the number of viewpoints or resolution should be prioritized. Currently, ISO (International Organization for Standardization) is discussing optical measurement methods for naked eye-type displays. And I think it is very important to establish a method to reasonably evaluate displays and determine required performances from the viewpoint of ergonomics.
Though many companies have announced a variety of naked eye-type 3D displays, consumers will find it difficult to understand the merits and features of each display if nothing is done. Because there is no common measurement method, technology development has been slowed. We have to develop evaluation standards that can be shared by consumers and engineers and can help advance technologies.
Do displays using glasses and naked eye-type displays have different application fields?
In the case of large displays, I think it is OK to require users to wear glasses. However, there are not many users who wear glasses to view small 3D displays such as mobile displays. So, small-size 3D displays should be naked-eye types.
What do you think is the problem of naked eye-type displays? How do you think it can be solved?
The issues of naked eye-type displays are the expansion of viewing angle, the prevention of the deterioration of resolution, the determination of the optimal number of viewpoints and the optimal width of each viewpoint, and so forth. To solve those problems, researchers are now working on head tracking and multi-viewpoints.
Though head tracking enables to make best use of the resolution of flat-panel displays, there were many problems in sensing, such as the detection of human heads, when I was working on it 10 years ago. Because sensing technologies such as facial recognition technologies have been drastically improved since then, it is now important to develop a method to make a display track humans.
In regard to multi-viewpoints, the resolution of a display is divided by the number of viewpoints, and the number of viewpoints and the width of each viewpoint are very important factors. On this point, we announced a method to optimize them based on the relation between the width of each viewpoint and the representation of 3D images in 2008.
Do you think 3D displays will be used for more types of devices in the future?
As I said earlier, we target naked eye-type displays at mobile devices such as mobile phones, portable game consoles and digital cameras.
Is there any issue that should be solved by the entire industry?
I think we should address the safety issue. To ensure the safety of 3D displays, it is necessary not only to improve the performance of displays but to be careful in making contents. Though 3D Consortium has already issued a guideline for the safety of 3D displays, we should keep improving the guideline based on various experiments and know-how and disseminate it in the entire industry.
As for optical measurement methods for displays, ISO and IEC (International Electrotechnical Commission) have already started to establish international standards. It is not possible to provide safe 3D displays to consumers without establishing a method to accurately evaluate the performance of displays. Since we currently do not have a standardized evaluation method, it is necessary for the industry to develop one as soon as possible.
By Hideo Anpo, Tech-On
