WO2009090599A1

WO2009090599A1 - Method for view rendering on an autostereoscopic display and an autostereoscopic display

Info

Publication number: WO2009090599A1
Application number: PCT/IB2009/050118
Authority: WO
Inventors: Waltherus A. H. Roelen
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2008-01-15
Filing date: 2009-01-13
Publication date: 2009-07-23
Also published as: TW200951490A

Abstract

Method for view rendering on an autostereoscopic display and autostereoscopic display wherein the view assignment for subpixels of the display is adapted such that for at least two view cones of different order, the respective view cones originating from two horizontally separated devices, the view cones overlap in a particular region in front of the autostereoscopic display, such that at a location within the region the first of the two view cones contributes a first color component for a particular view, and the second view cone contributes a second color component for the particular view, and wherein the view assignment within the spatial regions contributing to the first cone and the second cone differs.

Description

Method for view rendering on an autostereoscopic display and an autostereoscopic display

FIELD OF THE INVENTION

The present invention relates to a method for view rendering on an autostereoscopic display and an autostereoscopic display.

BACKGROUND OF THE INVENTION

For some time various companies have been actively developing autostereoscopic displays suitable for rendering three-dimensional (3D) imagery. Autostereoscopic devices can present viewers with a 3D impression without the need for special headgear and/or glasses. Autostereoscopic displays generally provide different views for different viewing angles. In this manner a first image can be generated for the left eye and a second image for the right eye of a viewer. By displaying appropriate images, i.e. appropriate from the viewpoint of the left and right eye respectively, it is possible to convey a 3D impression to the viewer. A variety of techniques are used to generate images for autostereoscopic displays. For example multi-view images can be recorded using multiple cameras wherein the position of the respective camera corresponds with the respective viewpoint of each respective view.

US patent 6064424 discloses and autostereoscopic display that use a film of slanted lenticular elements positioned on the display.

In the paper "Varrier Autostereoscopic Virtual Reality Display", by Sandin, D., et al as published in ACM Transactions on Graphics, Proceedings of ACM SIGGRAPH 2005, vol 24, no 3, an approach is presented to combine multiple stereoscopic displays into a single large 3D stereoscopic display. The Varrier paper discloses a display system that requires head tracking and which involves a complex registration scheme for registration of physical and computational parameters of the system.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method of view rendering on an autostereoscopic display wherein the 3D perception for viewers in a region in front of the autostereoscopic display are substantially improved.

This object is realized in that the view assignment for subpixels of the display is adapted such that at least two view cones of different order, the respective view cones originating from two horizontally separated spatial regions on the display, match in a particular region in front of the autostereoscopic display, such that at a location within the region in front of the autostereoscopic display, the first of the two view cones contributes a first color component for a particular view, and the second view cone contributes a second color component for the particular view, and the view assignment in the spatial regions contributing to the first cone and the second cone differs.

The present invention can through different assignments of views for different spatial regions providing viewing cones create sweet-spots in front of the autostereoscopic display such that multiple viewers within one or more sweet-spots of the display can perceive different images with their left eye and different images with their right eye resulting in a 3D impression.

Through the use of a shifted view assignment for the first and the second cone that are horizontally spaced apart, the first cone can contribute a color component to a pixel at a certain position in front of the display, and a second cone can contribute a further color component for the pixel at the same position in front of the display.

The above is particularly useful for wide autostereoscopic display, wherein the display covers a large part of the field of view of a user. In such a screen several horizontally spaced apart cones can contribute one (or optionally more) color components to a particular location in front of the screen. Due to the wide nature of the screen compared to the viewer distance to the screen, this principle can be exploited to provide a sound 3D impression in a region at a particular distance to the screen.

Here a screen is considered wide if the autostereoscopic display covers a viewing angle of an observer that is more than 60 degrees.

The present invention can be implemented for a single wide-viewing angle autostereoscopic display, or may alternatively be implemented for an array comprising at least two horizontally adjacent autostereoscopic displays, but preferably comprise at least three horizontally adjacent autostereoscopic displays.

Although above it is indicated that a single color component is contributed by a cone, this off course is dependent on the autostereoscopic display design. In fact a particular viewing cone may provide more than one color component. However the invention uses the fact that different viewing cones together contribute parts of a view.

BRIEF DESCRIPTION OF THE DRAWINGS These and other aspects of the invention will be further elucidated and described with reference to the drawing, in which

Fig. 1, shows a lenticular lens array in front of a (sub)pixel based display, Fig. 2, shows an example of subpixel assignment for an eight view 3D display, Fig. 3, shows an example of a simulation of how subpixels assigned to view 2 are visible through an optical system,

Fig. 4, shows the application of the invention in a planar setup of multiple multi-view based autostereoscopic displays,

Fig. 5, shows a top view of a multi-view display system according to the present invention wherein viewing cones originating from the respective displays are shown, Fig. 6, shows a perspective view of a multi-view display system according to the present invention,

Fig. 7, shows an application of the invention on a single wide autostereoscopic display, Fig. 8, shows the application of the invention in a non-planar setup of multiple multi-view based 3D displays,

Fig. 9, shows the optimal viewing area and optimal viewing distance line.

DETAILED DESCRIPTION OF EMBODIMENTS Figure 1 , shows a lenticular lens array in front of a (sub)pixel based display to form an autostereoscopic display device. Arrays of lenticular lenses can be used to direct the light from (sub)pixels in discrete directions.

Autostereoscopic display devices as disclosed in US patent 6064424 which may use a film of slanted lenticular elements positioned on the display. The lenticular film effectively directs light originating from (sub)pixels of a display panel into a variety of directions; thus providing a so-called viewing cone.

It is particularly noted that the subpixels as indicated by 105 are separated from the lenticules 110. As a result multiple viewing cones may be observed, wherein subpixels from different regions behind the lenticular contribute to a cone. The lenticular-based optical system in combination with the weaved pixel- layout (see Figure 2) provides a display panel where from each area on the display-screen a number (here nine) views is emitted in discrete directions. A set of views (here nine) is called a viewing cone. These cones are horizontally repeated. The optical system is designed in such way that cones from different areas on the screen coincide at the optimal viewing distance. A viewer at (approx.) the correct distance sees a pair of views and thus has may obtain an (auto- ) stereoscopic impression.

Since cones from different display areas coincide at the optimal viewing distance one consistent 3D image can be perceived from the total screen area. Since cones are repeated horizontally a consistent 3D (auto stereoscopic) perception can be obtained in many positions in front of the display

Figure 2, shows an example of subpixel view assignment for an eight view 3D display, this particular view assignment is intended for use with an autostereoscopic display device having a slanted lenticular film. The lenticular film is overlaid over the views and subsequently the lenticular film directs the light from the subpixels is various directions.

By assigning specific view information to the specific sub-pixels that are emitting their light in the same direction through the optical system a total view can be directed. In order to achieve this, the views assigned to the individual sub-pixels are 'weaved', such that in combination with a slanted lenticular placed on top it provides a proper image to the viewers. As presented in US patent 6064424 which discloses and autostereoscopic display that use a film of slanted lenticular elements positioned on the display.

Figure 3, shows an example of a simulation of how subpixels assigned to a particular view 2 are visible through an optical system comprising a lenticular. When combining multiple multi-view displays in one setup in order to create one 3D perception of the combine image a problem occurs.

Since multi-view displays have discrete viewing cones with a set of views forming subsequent stereoscopic pairs a combination of the position and angle of the viewer in respect to the display presents the viewer a different 3D perception (perspective). When using multiple multi-view displays the viewer has a different orientation and position in respect to every screen and thus sees different perspectives on each display not resulting in one consistent 3D perception.

The present invention proposes to create one multi-view perception from multiple screens the viewing cones and the corresponding views have to be matched. Figure 4, shows how a primary viewing cone comprising 9 views originates from a first display device 405 and may be perceived by the viewer 410. Apart from the primary viewing cone the lenticule positioned in front of the viewer 410 also provides two secondary cones, adjacent to the left and right of the primary cone. In addition a tertiary viewing cone is provided to the immediate left of the left secondary cone and to the immediate right of the right secondary cone.

Figure 4, further illustrates how the primary viewing cone of the first display device 405 is made to overlap with the tertiary viewing cone of the second, adjacent, display device 405'. This overlap as indicated in Figure 4 is obtained by a proper choice of view assignment for the second display device 405'. This alignment effectively is obtained by assigning views for the tertiary cone of the second display device to match with the primary cone of the primary device.

By assigning specific view information to the specific sub-pixels that are emitting their light in the same direction through the optical system a total view can be directed. In order to achieve this the views assigned to the individual sub-pixels are 'weaved', such that in combination with a slanted lenticular placed on top, provide a proper image to the viewers. As presented in US patent 6064424 which discloses and autostereoscopic display that use a film of slanted lenticular elements positioned on the display.

Figure 5, shows a top-view of a graphic rendition of a 3 dimensional scene wherein views from the respective displays are weaved. As can be seen the secondary viewing cone 520 of the primary display 505 is assigned views that match those of the secondary viewing cone of a further display 515.

Figure 6, shows a perspective view of the graphic rendition of the three dimensional scene from Figure 5, wherein the viewing cones are rendered and it is illustrated how the respective views of the respective cones overlap at the location where the viewers are positioned.

Figure 7, shows an application of the invention on a single wide autostereoscopic display.

The lenticular-based optical system in combination with the weaved pixel- layout provides a display panel where from each area on the display-screen a number (e.g. nine) views are emitted in discrete directions. A set of views (in this case nine) is called a viewing cone. These cones are horizontally repeated. The optical system is designed in such way that cones from different areas on the screen coincide at the optimal viewing distance. A viewer at (approx.) the correct distance sees a pair of views and thus may obtain a (auto-) stereoscopic perception.

Since cones from different display areas coincide at the optimal viewing distance one consistent 3D image can be perceived from the total screen area. Since cones are repeated horizontally a consistent 3D (auto stereoscopic) impression can be obtained in many positions in front of the display

The figure illustrates that by matching the higher order (here tertiary order) viewing cone with the primary viewing cone, the viewer of the primary viewing cone may also obtain a three-dimensional impression based on the subpixels in the spatial region providing the tertiary order cone.

Figure 8, shows how the present invention may also be used to weave non- planar views. Due to the fact that the views in the respective cones are narrow compared to the distance to the screen, the effect of the changed angle of the autostereoscopic display with respect to the viewer in Fig. 8 is limited. Figure 9, shows the optimal viewing area and optimal viewing distance line.

As can be seen from the figure multiple people can have a correct and consistent 3D viewing experience using an arbitrary set of multi-view displays.

By combining first and n-th order viewing cones the number of viewing areas is increased. There is no need for the production of special 3D multi-view based displays to build up a set of displays generating one 3D perception. The possibility to change the rendering in such way that views and viewing cones match is sufficient.

Using a combination of multiple 3D displays (multi-view) a larger display area can be achieved resulting in a more immersive perception of the displayed scenery. The combination of multiple displays further results in a higher overall resolution of the 3D images displayed.

The directions of the viewing cones of the different displays are matched by changing the rendering. In this way one can also direct the viewing cones of all displays to have optimal 3D perception in specific positions in respect to the set of 3D displays. The present invention may be used in conjunction with an autostereoscopic display for use in displaying still images or for use in displaying video sequences. It should be noted that once a proper view assignment is available the actual rendering of still and video images as such does not differ. It will be clear to the skilled person that an image processing device in accordance with the present invention can be constructed using either hardware components in the form of e.g. an Application Specific Integrated Circuit (ASIC) or multiple ASICs, or using a programmable platform comprising one or more digital signal processors or even general purpose processors. Combinations of the aforementioned implementations forms are also envisaged. The skilled person will be able to construct such an implementation based on constraints such as clock frequency, the available computational capacity on a programmable platform and/or other constraints well known to those skilled in the art of designing systems for video processing. It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims.

In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. It will be clear that within the framework of the invention many variations are possible. It will be appreciated by persons skilled in the art that the present invention is not limited by what has been particularly shown and described hereinabove. The invention resides in each and every novel characteristic feature and each and every combination of characteristic features. Reference numerals in the claims do not limit their protective scope. Use of the verb "to comprise" and its conjugations does not exclude the presence of elements other than those stated in the claims. Use of the article "a" or "an" preceding an element does not exclude the presence of a plurality of such elements.

Claims

CLAIMS:

1. Autostereoscopic display (705) wherein the view assignment (200) for subpixels of the display is adapted such that for at least two view cones of different order, the respective view cones originating from two horizontally separated devices, the view cones overlap in a particular region in front of the autostereoscopic display, such that at a location within the region the first of the two view cones contributes a first color component for a particular view, and the second view cone contributes a second color component for the particular view, and wherein the view assignment within the spatial regions contributing to the first cone and the second cone differs.

2. Autostereoscopic display (705) of claim 1, wherein the autostereoscopic display is implemented as an array of autostereoscopic display devices comprising at least two horizontally adjacent autostereoscopic display devices (405, 405').

3. Autostereoscopic display (705) of claim 2, wherein the at least two horizontally adjacent autostereoscopic display devices (805, 805') are positioned in a non- planar manner.

4. Autostereoscopic display of claim 1, wherein one of the at least two view cones is a primary order view cone.

5. Autostereoscopic display of claim 1, wherein the autostereoscopic display covers a viewing angle of an observer that is more than 60 degrees.

6. Method for view rendering on an autostereoscopic display, wherein the view assignment for subpixels of the display is adapted such that for at least two view cones of different order, the respective view cones originating from two horizontally separated spatial regions on the display, overlap in a particular region in front of the autostereoscopic display, such that at a location within the region in front of the autostereoscopic display, the first of the two view cones contributes a first color component for a particular view, and the second view cone contributes a second color component for the particular view, and the view assignment within the spatial regions contributing to the first cone and the second cone differs.

7. Method of claim 6, wherein the autostereoscopic display is implemented as an array of autostereoscopic display devices comprising at least two horizontally adjacent autostereoscopic display devices.

8. Method of claim 7, wherein the at least two horizontally adjacent autostereoscopic display devices are positioned in a non-planar manner.

9. Method of claim 6, wherein one of the at least two view cones is a primary order view cone.

10. Method of claim 6, wherein the autostereoscopic display covers a viewing angle of an observer that is more than 60 degrees.