DE19753040A1 - Method of generating 3D images for visual 3D perception based on virtual or real images for 3D vision in 3D video, 3D TV or 3D multimedia - Google Patents

Abstract

The method involves using a rastered, computer-controlled and addressable light generating structure, e.g. an array of micro-optical components, e.g. a microlaser array or a microfiber array with fiber-coupled, computer-controlled light sources of different wavelengths or spectral distribution to represent a real or computed 3D object. The 3D image information exists in the form of a digital data set as 3D point cloud. A first objective with a telecentric output is associated with the light generating structure and a second micro-optical array, which is connected to a computer-controlled linear scanner, which is further associated with a second objective for generating virtual or real images. An Independent claim is also included for an arrangement for carrying out the method.

Description

1. Technical field of application

The technical field of application is the creation of 3D images for the visual 3D perception based on virtual or real images for the 3D rendering. It is clear that the 3D effect is only due to the extremely complex Interaction of technical components and the visual sense of the contemplating subject arises. In extreme cases, this does not have to be human but it can also be an animal.

This is generally about the 3D vision in 3D video or 3D TV area, or around 3D multimedia applications of the well-known Usage. For example, 3D films can be played back. Also the rendering of calculated 3D computer graphics on the basis the display of virtual or real 3D images is possible. The image data are preferably in the form of digitally stored data, for example in Shape of a point cloud. The 3D impression can be colored, and it can moving objects and scenes can also be displayed.

2. State of the art

The generation of virtual or real 3D images (in the conventional Use of language) for visual perception for 3D rendering is based on State of the art in the generation of two differentially different images for the left and right eye. For this type of consideration is in many cases two-eyed goggles are required, which are either two different Color filter for the two eyes or two different polarizers having.

There are also shutter glasses that alternate a picture for the right and the left Clear eye. There are also glasses in which there is a separate one in each eye canal Image is generated. Although the quality of this 3D image reproduction can be very good, However, this technique is rejected and experienced for often very subjective reasons So far, there has been very little acceptance, even in domestic use. An exception is made represents the medical surgical technique. Here, for example, the surgeon wears one Special glasses, called cyber glasses, see VDI News from November 14, 1997, No. 46, P. 3.

Another option is to determine the position of the eye axes human viewer using electronic camera technology (Headfinder, Trucking System) and so from a common screen interface a different image for the left and the right eye using special technology. However, this is only possible for one observer  and is certainly a good technical possibility, but closes, for example 3D family television with multiple viewers.

Ultimately, based on the technique proposed by Nipkow, there is one Display of 3D images using a shutter. This is done in real time Perspective center of an image is constantly changing and at the same time only one Gap of the image released for viewing. The shutter can be mechanical or be built optoelectronic. In any case, it is only split by column Approval of the picture the picture is quite faint, since only a very small part of the Light energy reaches the observer.

The object of the invention is to provide a method and an apparatus which no laser radiation and / or special glasses required and the depth of field and improve the resolution.

According to the invention, this is achieved by the method and device according to the Claims 1 and 2. Advantageous refinements are in the subclaims specified.

3. Problem solved with the invention

The invention solves the task of generating virtual and real 3D images for the visual perception of the 3D shape of technical and natural bodies in the Space as well as complete scenes also on film or theater stages. Furthermore is based on the rendering of calculated 3D computer graphics the presentation of virtual or real 3D images, which is in the form of a There is a 3D point cloud.

Moving and colored 3D scenes can have a very high depth impression and with high lateral resolution in the form of virtual or real 3D images for the visual perception can be represented as 3D objects. The application area is used in the home area for television, video film playback, but also for Computer applications, generally seen in the multimedia field.

Also in professional areas such as medicine, u. a. the surgical Surgical technology, aerospace, simulation technology in general, the computer-aided design (CAD) as well as other C-techniques is the Task of 3D representation of data solved. In the scientific Research can use real data or models, such as Molecular representations are presented spatially. The same applies to DNA sequences in molecular biology and the modeling of genes in genetic engineering.

The possibility of spatial representation of mathematical functions in Education and research should also be mentioned, but also general School education. Not to forget the presentation of art objects and Fashion items. The possibilities to support the Orientation of people and computer-controlled objects in the room.  

The 3D images can be viewed as virtual images, for example by the viewer sits like in front of an umbrella and looks into an apparatus.

But also the generation of real images for 3D large-format reproduction for a larger audience is feasible. This is comparable to a 3D laser show.

4. Improvements and advantages achieved compared to the prior art

The application of the invention increases the depth display area or Depth impression and the image quality, including the brilliance and the point density the playback of 3D images.

The energy used for image generation is also optimally used. It arises a bright picture. The observation of the 3D image is basically by more than one Possible person. It is not necessary to wear special glasses.

On the use of directional laser radiation, as with laser projectors necessary, can be dispensed with.

5. Principles of the solution

The main features of the method and the arrangement for generating virtual and real 3D images together for visual perception described.

The process of creating virtual and real 3D images for the visual Perception is that by means of a screened, computer-controlled and thus addressable, light-giving structure that is basically voluminous or can be flat, for example an array of micro-optical Components such as a microlaser array or a microfiber array with fiber-coupled, computer-controlled light sources of different wavelengths, or spectral distribution a preferably planar self-illuminating partial image, in the sense of a for example, a perpendicular cut through a real or calculated one 3D object, i.e. as a boundary curve. However, it is also possible, a separate light source with a computer-controlled micro- Pair mirror array (Direct Mirror Device = DMD). The information about that 3D image to be displayed is in the form of a digital data record, for example in Form a point cloud.

This preferably planar and self-illuminating partial image on the array micro-optical components, which is also referred to as the first array, is through a first lens with a telecentric output, i.e. telecentric on the side facing away from the first array, preferably as a real image pictured. There is a depth range for the sharpness, which is less Aperture of the lens on the telecentric side enlarged, but the Luminous intensity of the lens and thus the brightness also decreases. This  Depth range is based on the approval of an acceptable Circle of confusion diameter for the pixels to be observed, the ultimately from the overall optical structure and the visual Resolving power of the viewer. Basically, too this first array can be moved linearly, only this is considered technically not so viewed advantageously.

In the depth range of the focus of the telecentric lens, there is a Another array with micro-optical components, which continues to be the second Array is referred to with micro-optical and focusing, preferably passive components. The real image of the planar and self-illuminating partial image on the first array.

Furthermore, the second array with a preferably linear, computer-controlled precision scanner with a direction of movement in the direction of Axis of the telecentric lens connected.

The scanner has a precise linear guide system, preferably a Parallel spring system for the precise linear movement of the second array with one Stroke on the order of a few millimeters to a few centimeters, however in the depth range of sharpness in the telecentric beam path.

The parallel springs of the parallel spring system can advantageously be made of consist of highly elastic, single-crystalline silicon.

The scanner is located with the second array, preferably in a vacuum, and is with a high-performance linear motor with a computer-coupled Precision encoder coupled. This ensures that a precise linear pre Backward movement with, for example, 25 Hz stroke frequency corresponding to one predetermined target function can be achieved without an undesirable Damping in the sense of braking by the air can occur.

Cross movements of the array to the optical axis can be computer controlled be useful if, for example, the second array consists of several Partial arays exist, each with its own linear motor and so the mechanical breakages should be made practically invisible. So can a particularly large Aray surface, for example for large-screen projection being represented. The lateral movement is in the pixel addressing on the the first array is taken into account, so is included.

Depending on the current, computer-controlled position of the second Arrays, a sectional image of the 3D Structure shown as self-illuminating. For example, up to 1000 self-luminous and possibly colored sectional images per second on the first Array created. Depending on the technical possibilities and targets, one higher or lower slice frequency can be realized.

The array preferably consists of convex microlenses, i.e. with focusing ones Effect. The size of the microlenses is chosen so that each microlens for example corresponds to a pixel. By the computer-controlled  Precision movement of the second array at approximately 25 Hz and within the The focus range of the telecentric lens is created for the human Observer the impression of a 3D image in space. By arranging focusing microlenses with a relatively large numerical aperture, For example, 0.3 or more of each luminous pixel becomes a light beam radiated in a relatively large solid angle range.

This resulting picture already has a 3D character for the observer, however, has a relatively shallow depth and is relatively small. Therefore a second imaging lens with the aim of enlarging the image and the depth downstream. The image on the second lens array, which can be seen as an intermediate picture, in the sense of technical Optics can be done virtually or real through the second lens.

It is essential that the virtual image with both eyes at the same time is looked through the second lens. The second array points in relation to the second lens an intrafocal position with the limit case of the focal position. In the focal position of the second array, an observer is behind the second lens seen an image in infinity. With the linear movement of the Arrays in the direction of the second lens, the image is created as a virtual image, which, however, is viewed with the eye as in magnifying glass without a screen can be.

A real image is generated for the large image projection. This is where second array each in the extrafocal projection of the second imaging lens, which then represents a specially adapted projection lens. The real pictures arise in the direction of light behind this lens, for example in a fog wall or a cloud chamber. Also is a computer controlled three dimensional matrix with computer-controlled microreflectors conceivable in 25 Hz intervals different depths, i.e. in the room, are switched, for example so that the beam hits or can pass. This also creates a 3D impression for the observer. Veil walls are also for more artistic projects applicable.

In the virtual representation of the images, a monolithic can be used as the second lens Lens with a diameter well above the eye relief is used become. However, a lens as large as a Fresnel lens is better To design sections, including micro-sections, with regard to all Aberrations, including color errors, should be corrected. This size Lens has exactly one optical axis. Microlenses can also be used here Come into play. For example, even several microlenses in the direction of light be arranged one behind the other to correct the aberrations, or one Achieve optimization of the mapping.

When designing the large lens, all possibilities of optical know-how can mapping correction, including microstructuring for the Design of the lens can be used. Even the most modern micro Antireflection techniques can be used for the surface design of the lens become. Instead of such a high-tech lens can also be used for the second A multi-lens lens can also be used.  

The second lens can be coupled with an airtight recipient that the second array, the scanner, the inside of the second lens and the recipient walls enclose a vacuum already mentioned. Here the recipient has a transparent entrance window through which the light comes out the telecentric lens falls on the second array.

It is also conceivable that only the second array and the scanner are in a vacuum are located. However, this may result in the crooked rays that arise due to the large aperture of the microlenses result in an imaging problem Connection with the exit window then necessary. It can Astigmatism may occur.

When designing the second array, knowledge of dynamics can be used moving elastic body can be applied. The due to the no small Acceleration forces in the second array thanks to the high-performance linear motor occurring dynamic deformations of the second array, which from Plastic can exist in the creation of images for the 3D effect be included. Finally, the surface shape of the second array basically freely selectable. The shape of the array at a given time only has to be known exactly. So a curved shape is also conceivable, for example in the sense of mass and deformation minimization. This is possible because all points for the creation of the individual images are temporally and locally are addressable.

The second lens can also be a concave mirror, preferably in an off-axis Arrangement, be. The second array is in an on-axis Position. The combination of. Can also be used to correct aberrations Concave mirror and one or more lenses can be selected. This approach too can be designed with micro-optical components as a Fresnel mirror. It is only important here that this mirror has a single optical axis.

It is completely clear that 3D visualization involves data streams in gigabyte Range can go per second. Data reduction techniques can also be used be used.

The image capture of real 3D objects and scenes can be done with that of the same applicant with the priority of 5 November 1997 filed patent application entitled "Method and apparatus for generating a 3D point cloud for the Topometry (3D video measurement) or the 3D vision for multimedia applications, (DE- 197 49 974.0-31) "can be realized.

6. Embodiments

In FIG. 1, the apparatus for generating 3D images for visual 3D perception on the basis of virtual or real images is represented in the essential components. The computer-controlled photo series, which is created on an addressable light-emitting structure in the form of a laser diode matrix, is imaged by a first lens with a telecentric output.

The current image comes through a window onto a microlens array focusing microlenses. Each microlens represents a pixel, of which a wide open bundle due to the relatively large numerical aperture of the Microlenses, for example starting from 0.3. The microlens array is with one computer-controlled high-precision linear scanner connected. In the Laser diode matrix, the images are generated so that a real or virtual 3D Object is broken down into vertical sections and each on the surface of the 3D Object lying object points as pixels on the laser diode matrix being represented. The 3D object is available as a 3D point cloud. Taking into account the Image relationships through the 1st lens, the linear scanner with the Microlens array so that the real or virtual 3D object, its 3D Data set as a point cloud and thus as a template for controlling the Laser diode matrix was used in a geometrically similar sense.

For the unimpeded movement of the microlens array, the linear Scanner in a vacuum. This will hamper the highly dynamic No movement due to air friction.

The microlenses, which actually represent the pixels, of the microlens Arrays of outgoing beams are guided through a second lens. This is practically a large lens with a magnifying effect, for example one 400 mm focal length. From today's perspective, this large lens can be an absolute High-tech product incorporating all modern optical technologies represent with the aim of correcting the imaging of the microlens array below the real conditions of viewing the 3D image through a group of People from different depths of the living space.

Due to the intrafocal position of the microlens array, the technical optics a virtual image of one or more people can be perceived directly. Taking into account the entire illustration of the laser diode matrix to the eye of the observer or the Laser diode matrix and the linear scanner controlled so that on the one hand a 3D Impression arises that comes very close to the natural feeling. On the other hand are special, even extremely distorting effects, especially for multimedia Applications, doable.

Fig. 2 shows the formation of the virtual image from the viewer. When point A on the microlens array moves from position 1 to position 2, virtual points A1 'and A2' are created in the known manner, namely in different depths and sizes. In connection with the 3D point cloud, these imaging ratios must be converted into the control information for the laser diode matrix and the linear scanner. This creates the 3D impression of the viewer in connection with the visual 3D perception.

Claims (1)

Arrangement for generating an impression of 3D images for visual perception on the basis of virtual or real images for 3D vision in the 3D video or 3D television area or for 3D multimedia applications, characterized in that that by means of a rastered, computer-controlled and thus addressable, light-emitting structure, which can be basically voluminous or flat, for example an array of micro-optical components such as a microlaser array or a microfiber array with fiber-coupled, computer-controlled light sources of different wavelengths or spectral distribution preferably planar self-illuminating partial image, in the sense of, for example, an axis-perpendicular cut through a real or calculated 3D object, and the information about the 3D image to be displayed is available in the form of a digital data record as a 3D point cloud and a first lens with a telecentric output illuminating n structure is assigned and the second lens is assigned a second micro-optical array which is connected to a computer-controlled linear scanner, which in turn is followed by a second imaging lens for generating virtual or real images.