WO2011151044A2 - Method and arrangement for three-dimensional representation - Google Patents

Abstract

The invention relates to a method and arrangement for three-dimensional representation. The method comprises the following steps: Front or back projection of at least two views A (k) where k=1..n and n>1 onto a projection screen (1) in a chronologically sequential succession and in a defined resolution of pixels x (i, j) in lines i and columns j, chronologically sequential partial shading of each pixel x (i, j) on the projection screen (1) such that at any time (t) precisely one sub-surface T (k, i, j) of each pixel x (i, j) is visible, wherein during the projection of any fixed view A (k) the same sub-surface T (k, i, j) of a pixel x (i, j) always remains visible and the sub-surfaces T (k, i, j) where k=1..n for fixed pairs (i, j) and different k do not overlap on the projection screen (1), and displaying the sub-surfaces T (k, i, j) in at least n directions, so that from at least one viewing position a viewer sees at least some image information of various views with both eyes, whereby a three-dimensional perception is generated. A special embodiment relates to a high-resolution 2D representation.

Description

 Method and arrangement for spatial representation

The invention relates to the field of spatial representation, in particular the spatially perceptible presentation without aids for simultaneously multiple viewers, the so-called autostereoscopic visualization.

For some time, there have been approaches to the aforementioned subject area. A pioneer in this field was Frederic Ives, who introduced in GB1904-18672 a system with a "line screen" for the 3D representation, as well as in Sam H. Kaplan's "Theory of parallax barriers", Journal of SMPTE 59 , No 7, pp 11-21, July 1952 describes basic insights on the use of barrier screens for 3D imaging.

 For a long time, however, no comprehensive dissemination of autostereoscopic systems was achieved. Only in the 80s of the 20th century could a certain renaissance of the 3D systems be used due to the now available computing power and new display technologies. In the 1990s, the number of patent applications and publications on glasses-free 3D visualizations skyrocketed. Outstanding results have been achieved by the following inventors or providers:

In JP 8-331605, Ikeda Takashi et al. (Sanyo) a step barrier in which a transparent barrier element has approximately the dimensions of a color subpixel (R, G or B). With this technique, it was possible for the first time to partially redirect the vertical resolution of the loss of resolution in the horizontal direction that occurs in most autostereoscopic systems due to the simultaneous display of several views (at least two, preferably more than two views). The disadvantage here, as with all barrier methods, is the high loss of light. In addition, when the viewer moves sideways, the stereo contrast changes from nearly 100% to about 50%, and then increases again to 100%, resulting in a 3D image quality that fluctuates in the viewing space. Pierre Allio, with the teachings of US 5,808,599, US 5,936,607 and WO 00/10332, achieved a notable advancement in lenticular technology, also using sub-pixel based view partitioning. Another outstanding R & D result was filed for patent by Cees van Berkel (Philips) with EP EP 0791 847. Lenticular lenses inclined relative to the vertical are located above a display which also shows different perspective views. Characteristically n views are divided on at least two screen lines, so that in turn the loss of resolution from the horizontal is partially transferred to the vertical.

 Several milestones for autostereoscopy were justified by Jesse Eichenlaub with the documents US Pat. No. 6,157,424 and WO 02/35277 as well as numerous other inventions which, however, represent almost all 3D systems for only one observer and / or are often not manufacturable at acceptable costs.

 With DE 10003326 C2 Armin Grasnick et al. 4D-Vision advances barrier technology to two-dimensionally structured wavelength-selective filter arrays to create a 3D impression. However, the disadvantage here too is the greatly reduced brightness of such 3D systems compared with a 2D display.

 Armin Schwerdtner (Seereal Technologies) succeeded with WO 2005/027534 a novel technological approach for a full 3 D representation in all (usually two) views. However, this approach is associated with high adjustment effort and for larger screen diagonals (from about 25 inches) is extremely difficult to implement.

 Finally, Wolfgang Tzschoppe et al. (X3D Technologies) WO 2004/077839 A1, which relates to a brightness-enhanced barrier technology. Based on the approach of a step barrier of JP 08-331605 and DE 10003326, a special duty cycle of the transparent to the opaque barrier filter elements is presented here, which is greater than 1 / n with n the number of views shown. However, the embodiments and teachings disclosed in this document generally produce a severely limited perception of depth, since the stereo contrast is greatly reduced compared to approximately the teaching of JP 08-331605.

The invention is based on the object of providing a possibility for autostereoscopic display in order to achieve improved visibility for several viewers at the same time. Improved perceptibility is particularly, but not exclusively, an improved resolution simultaneously to understand the highest possible stereo channel separation. Appropriate methods and arrangements should be industrially feasible and at the same time present a spatial image to multiple viewers without the need for 3D glasses. Furthermore, a method for a particularly high-resolution two-dimensional representation is to be specified.

This object is achieved by a method for spatial representation, comprising the following steps

 Projection of at least two views A (k) with k = 1..n and n> 1 onto a projection screen in temporal sequential order and in a defined resolution of picture elements x (i, j) in lines i and columns j .

 temporally sequential, partial shading of each picture element x (i, j) on the projection screen so that exactly one sub-area T (k, i, j) is visible at each point in time t of each picture element x (i, j) Projection of any but fixed view A (k) always the same sub-area T (k, i, j) of a picture element x (i, j) remains visible and the sub-areas T (k, i, j) with k = 1. .n for fixed pairs (i, j) and different k do not overlap on the screen,

 Mapping of the sub-areas T (k, i, j) in at least n directions, so that from at least one viewing position a viewer sees with both eyes at least partially image information of different views, so that a spatial perception arises. Preferably, the temporal sequential, partial shading is realized by a flat light valve, for example a liquid crystal shutter, which is in the immediate vicinity of the projection screen and which temporally sequentially and in synchronization with the projection of the views A (k) the projected light only on partial surfaces T ( k, i, j).

In other words, the invention first makes use of the fact that some reflective imagers of projection technology are particularly fast, such as DMDs or fLCOS chips or the like. By virtue of such a fast reflective imager, in the first variant, the backside of a projection screen becomes temporally sequential Sequence of at least two views A (k) projected. According to the invention, a shutter or similar optical component is now arranged in front of or behind the projection screen, which leaves only a portion visible per pixel (in this case per full color pixel). This leaves the full resolution 5 visible at all times, but the pixels no longer show the full fill factor.

Alternatively, the temporal sequential, partial shading can be realized by an opto-mechanical system, for example, a rotating surface with a reflective surface, at least partially with an opaque pattern

ID which temporally sequentially and in synchronization with the projection of the views A (k), the projected light only on partial areas T (k, i, j) maps. Such a rotating surface can at the same time still the function of a color wheel (as usual with DLP projectors), so take over the modulation of the color. Furthermore, it may be useful, such an opto-mechanical system in two parts

For example, one mirror wheel would then modulate the color, and the other mirror wheel would ensure the patch imaging because of the opaque pattern. Both mirror wheels do not necessarily have to work at the same rotational frequency. The imaging of the partial areas T (k, i, j) in at least n directions is carried out, for example, by means of a lenticular screen, a lenticular screen, a parallax barrier, a holographic-optical element (HOE), a prism grid, a structured surface and / or a diffraction pattern , Other embodiments are conceivable.

Alternatively, it is possible that the imaging of the partial areas T (k, i, j) is carried out in at least n directions by means of dynamic optics, for example a time-varying lenticular grid. Of course, then a suitable synchronization with the different state modes of the lenticular must be ensured with the views per unit time views A (k).

D

 The object of the invention is also achieved by an arrangement for spatial representation, comprising

a projection unit for projecting at least two views A (k) with k = 1..n and n> 1 onto a projection screen in front or back temporally sequential sequence and in a defined resolution of pixels x (i, j) in rows i and columns j,

 Means for temporally sequential, partial shading of each picture element x (i, j) on the projection screen such that each picture element x (i, j) has exactly one face T (k, i, j) visible at each instant t, where in the projection of any but fixed view A (k), the same subarea T (k, i, j) of a picture element x (i, j) always remains visible and the subareas T (k, i, j) with k = 1..n for fixed pairs (i, j) and different k do not overlap on the screen,

 - Means for imaging the partial surfaces T (k, i, j) in at least n directions, so that from at least one viewing position, a viewer with two eyes at least partially sees image information of different views, so that a spatial perception arises.

 The subareas T (i, j) [but not T (k, i, j)] effectively correspond to pixels which are divided into k sections and which successively reproduce image information of different views A (k) at the different sections k.

The means for temporally sequential, partial shading consist of a flat light valve, for example a liquid crystal shutter, which is located in the immediate vicinity of the projection screen and which temporally sequentially and in synchronization with the projection of the views A (k) the projected light only on partial surfaces T ( k, i, j).

 If necessary, the light coming from the projection screen must be depolarized or generated unpolarized in order to pass through the shutter.

Alternatively, the temporal sequential partial shading means may consist of an opto-mechanical system, for example a rotating surface having a specular surface at least partially provided with an opaque pattern which is sequential in time and in synchronization with the projection of views A (FIG. k) images the projected light only on partial areas T (k, i, j). Such a rotating surface can at the same time still the function of a color wheel (as usual with DLP projectors), so take over the modulation of the color. Furthermore, it may be useful, such an opto-mechanical system in two parts Then, for example, one mirror wheel would modulate the color, and the other mirror wheel would ensure the patch imaging because of the opaque pattern. Both mirror wheels do not necessarily have to work at the same rotational frequency.

 5

 The means for imaging the partial areas T (k, i, j) in at least n directions consist, for example, of a lenticular screen, a lenticular screen, a parallax barrier, a holographic-optical element (HOE), a prism grid, a structured surface and / or a diffraction pattern , The ID period widths of the abovementioned optical elements can come into question from the pixel width of a partial area T (i, j) up to a multiple width of a partial area (i, j).

In special embodiments, the means for imaging the partial surfaces 15 T (k, i, j) in at least n directions can consist of a dynamic optical system, for example a time-varying lenticular grid. Of course, then a suitable synchronization with the different state modes of the lenticular must be ensured with the views per unit time views A (k). Advantageously, the projection beam path can be folded from the projection unit to the projection screen. Particularly advantageously, the means for folding the beam path contribute at least partially to the mode of action of the means for temporally sequential, partial shading and / or the mode of action of the means for imaging the partial surfaces T (k, i, j) in at least n directions.

5 The flat design of a screen based on the projection is e.g. using the teaching of WO9953375 or WO0163356 possible.

The projection screen may in special embodiments be a front projection screen, which luminesces for a few milliseconds. For this purpose, for example, a time-delayed (eg luminescent) screen could be used, so that the views are each emitted in other directions, that is, from which they come. In the representation of sufficiently many, eg 30, views A (k) unpleasant transitions of the views A (k) are reduced in head movement of the viewer. Finally, the application of the inventive concept also provides an arrangement for high-resolution two-dimensional representation, comprising

 a projection unit for projecting at least two pairwise disjoint subsets A (k) of one and the same picture with k = 1..n and n> 1 onto a projection screen in temporal sequential order and in a defined resolution of picture elements x (i , j) in lines i and

Columns j,

 Means for temporally sequential, partial shading of each picture element x (i, j) on the projection screen such that each picture element x (i, j) has exactly one face T (k, i, j) visible at each instant t, where in the projection of any, but fixed subset A (k) always the same

Subarea T (k, i, j) of a picture element x (i, j) remains visible and the subareas T (k, i, j) with k = 1..n for fixed pairs (i, j) and different k on do not overlap the screen,

 such that all the pairs of disjoint subsets A (k) of the image are temporally sequential visible to a viewer and thus a high-resolution

2 D image is displayed.

The subsets A (k) replace the views in this context. Advantageously, the subsets are each composed of strips of the same view AO, wherein in each case, for example, all nth columns of a view AO, starting at the kth column, are taken over into a subset A (k).

The invention will be explained below with reference to exemplary embodiments. The drawings show:

Fig. 1 shows the schematic structure for implementing the invention

 process

FIGS. 2 and 3 each show the schematic structure in the method according to the invention at two different points in time, and FIGS Fig. 4 is a diagram for forming an opto-mechanical system for use in the method according to the invention.

All drawings are not to scale. This concerns in particular also 5 angle dimensions, if available.

1 shows the schematic structure for implementing the method according to the invention.

 The exemplified inventive method for spatial representation

ID includes the following steps

 Rear projection of two views A (k) with k = 1... 2 from a projector 3 onto a projection screen 1 in temporal sequential order and in a defined resolution of pixels x (i, j) in rows i and columns j, temporally sequential , partial shading of each pixel x (ij) on the

15 projection screen 1, so that each pixel x (i, j) at each time point t respectively exactly one face T (k, i, j) is visible, wherein in the projection of any, but fixed view A (k) always the same partial area T (k, i, j) of a pixel x (i, j) remains visible and the partial areas T (k, i, j) with k = 1 ..n for fixed pairs (i, j) and different k on the projection screen 1

Do not overlap ZD,

 Imaging the partial areas T (k, i, j) in at least two directions so that from at least one viewing position a viewer 5 with both eyes L and R sees at least partial image information of the different views A (1) and A (2), so that a spatial perception arises.

 25

 The temporally sequential, partial shading is realized here by a flat light valve 2, for example a liquid crystal shutter, which is in the immediate vicinity of the projection screen 1 and which temporally sequentially and in synchronization with the projection of the views A (k) the 30 projected light only on Partial surfaces T (k, i, j) lets happen.

The imaging of the partial areas T (k, i, j) in at least n = 2 directions is carried out, for example, by means of a lenticular screen 4. Other embodiments are conceivable. FIGS. 2 and 3 each show the schematic structure in the method according to the invention at two different points in time. The circled number "1" or "2" indicates, respectively, that at the respective time the projector 3 reproduces the view 1 or 2 in full resolution. Accordingly, the light valve 2 is in

5 switched its columns so that each one half of a partial area T (i, j) is transparent, and the other is opaque. As a result, only a corresponding part of the surface sections T (i, j), and therefore the partial surfaces T (1, i, j) and T (2, i, j), respectively, with image information of the views A (1) and A (2). The imaging by means of the lenticular lenses 4 thus ensures that the

I0 view A (1) only the left eye L and the view A (2) reaches only the right eye R. The method according to the invention and, as it were, the drawings could also be modified for n> 2 views A (k) with k = 1..n.

Finally, Fig. 4 shows a scheme for forming an opto-mechanical

15 Systems 6 for use in the method according to the invention. In this case, the temporally sequential, partial shading instead of a light valve 2 by an opto-mechanical system 6 can be realized, for example, a rotating surface with a specular surface, which is at least partially provided with an opaque pattern, which temporally sequential and in synchronization with the 0 Projection of the views A (k) the projected light only on sub-areas T (k, i, j) maps. At the same time, such a rotating surface 6 can still assume the function of a color wheel (as is usual with DLP projectors), ie the color modulation, as in FIG. 4 by the letters R, G, B (for red, green and blue). indicated. This is achieved by the use of colored (R, G, B) reflecting mirrors or surfaces. At the same time as described above, the opaque pattern is still present, as indicated in Fig.4 in the lower portion of the red reflecting surface on the Farbspiegelrad 6. The pattern provides the appropriate partial shading in the sense described above. The color mirror wheel rotates at a rate synchronous with the timing of the image of the views A (k) so that each view A (k) is projected in true color.

The drawings can also be used analogously for the illustration of the arrangements according to the invention. Fig.1 would then be mutatis mutandis understood as an arrangement for spatial representation, comprising a projection unit 3 for projecting at least two views A (k) with k = 1..n and n> 1 onto a projection screen 1 in chronologically sequential order and in a defined resolution of picture elements x (i, j) in FIG Rows i and columns j,

 Means 2 for temporally sequential, partial shading of each picture element x (ij) on the projection screen 1, so that exactly one sub-area T (k, i, j) of each picture element x (i, j) is visible at each time t, where in the projection of any but fixed view A (k) always the same subarea T (k, ij) of a picture element x (i, j) remains visible and the subareas T (k, i, j) with k = 1. .n for fixed pairs (ij) and different k on the projection screen 1 do not overlap,

 Means 4 for imaging the partial surfaces T (k, i, j) in at least n directions, so that from at least one viewing position, a viewer 5 sees with both eyes at least partially image information of different views, so that a spatial perception arises.

 In order to avoid repetition, the corresponding explanation of the other drawings is omitted here.

The advantages of the invention are versatile: in particular, a resolution of the SD image and a good brightness is made possible. Furthermore, the imager of the projection system (imagers such as DMD or LCOS) does not need to be moved, and apart from a possible color wheel, there are also generally no moving parts.

Claims

claims 1. A method of spatial representation, comprising the following steps  Projection of at least two views A (k) with 5 k = 1..n and n> 1 onto a projection screen (1) in temporally sequential projection  Sequence and in a defined resolution of pixels x (i, j) in rows i and columns j,  temporally sequential, partial shading of each picture element x (i, j) on the projection screen (1), so that from each picture element x (i, j) to each I0 time t each exactly a partial area T (k, i, j) is visible, wherein at the  Projection of any but fixed view A (k) always the same sub-area T (k, i, j) of a picture element x (ij) remains visible and the sub-areas T (k, i, j) with k = 1..n for fixed pairs (i, j) and different k do not overlap on the screen (1),  15 - imaging of the partial surfaces T (k, i, j) in at least n directions, so that from at least one viewing position a viewer sees with both eyes at least partially image information of different views, so that a spatial perception arises. 0 2. The method of claim 1, characterized in that the temporally sequential, partial shading is realized by a flat light valve, such as a liquid crystal shutter, which is in the immediate vicinity of the projection screen (1) and which temporally sequential and in synchronization with the projection the views A (k) allow the projected light 5 to pass only on sub-areas T (k, i, j). 3. The method according to claim 1, characterized in that the temporally sequential, partial shading is realized by an opto-mechanical system (6), for example, a rotating surface with a mirror surface which is at least partially provided with an opaque pattern, which in time sequentially and in synchronization with the projection of the views A (k), projecting the projected light only onto sub-areas T (k, i, j). Method according to one of the preceding claims, characterized in that the imaging of the partial surfaces T (k, i, j) in at least n directions by means of a lenticular screen, a lenticular screen, a parallax barrier, a holographic-optical element (HOE), a prism grid, a structured surface and / or a diffraction pattern is performed. Method according to one of claims 1 to 3, characterized in that the imaging of the partial surfaces T (k, i, j) in at least n directions by means of a dynamic optics, for example, a time-varying lenticular grid, is performed. Arrangement for spatial representation, comprising  - A projection unit (3) for front or back projecting at least two views A (k) with k = 1..n and n> 1 on a projection screen (1) in temporal sequential sequence and in a defined resolution of pixels x (i , j) in rows i and columns j,  Means (2) for temporally sequential, partial shading of each picture element x (i, j) on the projection screen (1) such that each picture element x (i, j) has at each point in time t exactly one face T (k, i, j) is visible, wherein in the projection of any but fixed view A (k) always the same subarea T (k, i, j) of a picture element x (i, j) remains visible and the faces T (k, i , j) with k = 1..n for fixed pairs (i, j) and different k on the projection screen (1) do not overlap,  Means (4) for imaging the partial surfaces T (k, i, j) in at least n directions, so that from at least one viewing position a viewer (5) sees with both eyes at least partially image information of different views, so that a spatial perception arises. Arrangement according to claim 6, characterized in that the means (2) for temporally sequential, partial shading consist of a flat light valve, for example a liquid crystal shutter which is in the immediate vicinity of the projection screen (1) and which is sequential in time and in synchronization with the Projection of the views A (k) allows the projected light to pass only on partial areas T (k, i, j). 8. Arrangement according to claim 6, characterized in that the means (2) for temporally sequential, partial shading consist of an opto-mechanical system (6), for example, a rotating surface with a reflective surface, at least partially with an opaque pattern which, in time sequential and in synchronization with the projection of the views A (k), images the projected light only on partial areas T (k, i, j). ID 9. Arrangement according to one of claims 6 to 8, characterized in that the means (4) for imaging the partial surfaces T (k, i, j) in at least n directions consist of a lenticular screen, a lenticular screen, a parallax barrier, a holographic optical element (HOE), a prism grid, a textured surface and / or a 15 diffraction pattern. 10. Arrangement according to one of claims 6 to 8, characterized in that the means (4) for imaging the partial surfaces T (k, i, j) in at least n directions from a dynamic optics, such as a time-varying G lenticular consist. 1 1. Arrangement according to one of claims 6 to 10, characterized in that the projection beam from the projection unit (3) to the projection screen (1) is folded. 5  12. Arrangement according to claim 1 1, characterized in that the means for folding the beam path at least partially contribute to the operation of the means (2) and / or the means (4). D 13. Arrangement according to one of claims 6 to 10, characterized in that the projection screen (1) is a front projection screen, which luminesces for a few milliseconds. 14. Arrangement for high-resolution two-dimensional representation, comprising a projection unit (3) for projecting at least two pairwise disjoint subsets A (k) of one and the same picture with k = 1..n and n> 1 onto a projection screen (1) in temporal sequential order and in one defined resolution of pixels x (i, j) in rows i and columns j,  Means (2) for temporally sequential, partial shading of each picture element x (i, j) on the projection screen (1) such that each picture element x (i, j) has at each point in time t exactly one face T (k, i, j) is visible, wherein in the projection of any, but fixed subset A (k) always the same sub-area T (k, i, j) of a picture element x (i, j) remains visible and the sub-areas T (k, i , j) with k = 1..n for fixed pairs (i, j) and different k on the projection screen (1) do not overlap,  so that in time sequentially for a viewer (5) all pairwise disjoint subsets A (k) of the image are visible and thereby a high-resolution 2D image is displayed.