TWI278225B - Multiple views of stereoscopic image display scheme - Google Patents

Abstract

A multiple views of stereoscopic image display scheme is disclosed, wherein one set of multiple-views images (more than two) are individually and separately arranged/processed to form a frame image, the frame image actually is the arrangement of the thumbnail of the multiple-view image, digital image compression processing can be proceeded onto such a frame image by the image compression technology (such as MPEG1, MPEG2, ...), which is convenient for the digital storage of stereoscopic image data. Otherwise, if digital image compression processing is proceeded by the last synthesized stereoscopic image, the synthesizing format will be destroyed and even results in losing the stereoscopic effect. In addition, to solve the real-time playing efficiency problem of multiple-views stereoscopic image, we can transmit the arranged frame of each thumbnail in the multiple-views stereoscopic image to a stereoscopic image synthesizer, so as to increase the operating efficiency of computer. The color code frame is used to inform the stereoscopic image synthesizer for synthesizing the stereoscopic image, and to output to a flat panel display such as LCD monitor. At this time, if a lenticular lens is installed on the flat panel display, which is proper for the flat panel display to display with the specification of the multiple-views stereoscopic image, it is not necessary to wear LC shutter glasses for us to watch the stereoscopic image frame autostereoscopically.

Description

1278225 IX. Description of the Invention: [Technical Field] The present invention relates to an autostereoscopic stereoscopic image display processing method, and more particularly to a Multiple Views of Stereoscopic Image Display Scheme. This multi-view image is image-isolated (7)(10)

Arrangement) 'A color recognition bar code (c〇i〇rc〇des) is used to tell the stereoscopic image synthesizer to synthesize the separated images of the facet for an interlaced display on the Lenticular Lens. Multi-view stereoscopic image format allows users to wear stereo glasses without

For flat panel displays such as LCDs with Lenticular Lens

Monitor) to view the stereoscopic image. [Prior Art] Humans have three-dimensional vision because they have two eyes, and each eye can view the same scene and the parallax generated by different angles of view is mainly present in the visual nerve of the brain. In the past, in the environment of personal computer, we were able to wear stereo glasses (LCShutterGlasses), and watch the interlaced scanning mode with the vertical sync signal (VerticalSync) control to make the left and right eye images alternately switch display. The stereoscopic image basically consists of placing the even-numbered strips in the scan line on the left (or right) eye image, and the odd-numbered strips placing the right (or left) eye image, in turn alternately switching the display, when the fluorescent screen When the even number is displayed, the left eye of the user can only see the image of the left eye. At this time, the stereo glasses will automatically match the interlaced display mode to cover the right eye; similarly, when the fluorescent screen displays an odd number of lines, the right eye Only the image of the right eye can be seen, so the stereo glasses will automatically match the staggered display mode to cover the left eye. In this way, the left and right eye images are sent independently to the left and right eyes, and faster than the speed of the human visual persistence. The switching display shows that the stereoscopic (9) (10) 3D image can be clearly seen. When viewing the stereoscopic images of two views (Views), most of the viewers need to wear stereo glasses to view the stereoscopic images. However, such a way of calling stereo glasses often makes the viewer feel uncomfortable, and at the same time, wearing a pair of glasses, it is inconvenient to wear stereo glasses to view the stereoscopic image L, it is necessary to use another type without need to shout stereo Glasses to view stereoscopic images instead of 'Multiple Views' stereo imagery with Lenticular Lens display is the most cost-effective technology at present, common in stationery, gifts, toys, bag 278225 Printed goods, etc.; although multiple views (Lenticular Lens) are widely used in printing technology, it has been a tradition for more than a decade. Technology; however, technology is constantly evolving, and image display technology has gradually been replaced by the current compact and bulky cathode ray tube (CRT) display by the current light and thin flat panel display (jqatpanel display), especially The LCD monitor is the most representative. Therefore, the stereoscopic image that we want to display will also be transferred to the LCD screen display. What it wants to display is not only the general stereoscopic image, but the animation (Motion) and the Still Image interactive Real Time stereoscopic shirt. Handled interleaved (interleaved or interweaved) multi-view stereo image compression format (such as MPEG1, MPEG2, ...), restored and played back after decompression, because the current compression technology (such as, ...) is destructive compression It can be accepted for planar images with less precision and accuracy alignment, but it is not suitable for the compression of inter-accurate (interieavecj or interweaved) multi-view stereo images with precision and accuracy. If we still want to use the existing compression technology (WMPEGI, MPEG2, ...) to deal with the compression of interlaced (interieavec^interweaved) multi-view stereo images, then we must have special skills and methods; On the other hand, for interactive stereoscopic images that require instant “T^ime” processing, the synthesis of multi-view stereoscopic images (especially multi-view stereoscopic image synthesis with super-resolution) will consume quite a lot of computers. The processing time, so it is necessary to arrange the synthesis work of the multi-view stereo image to a dedicated Synthesizer to process 'to improve the working efficiency of the computer. Combining the above two reasons: solving the problem of digital compression distortion of multi-view stereo image and improving the working efficiency of the computer, so I will propose the invention as a countermeasure. SUMMARY OF THE INVENTION The main object of the present invention is to solve the above-mentioned problems of stereoscopic images by stereoscopic glasses, and to solve the problem of digital compression distortion of multi-view stereo images and improve the working efficiency of the computer. (Multiple Views) Stereoscopic image display processing method allows viewers to view high-efficiency, high-quality, distortion-free stereoscopic images without wearing stereo glasses. I. 6 1278225 For the above purposes, the new method utilizes a multi-view (Multiple Views) image, a View Arrangement, a Color Codes frame, and a stereo A multi-viewer (Synthesizer) consists of a set of multiple view images (two or more), and a separate frame (View Arrangement) to form a frame image. In fact, it is the arrangement of the various thumbnails of the multi-view image, and uses the ColorCodes frame as a communication tool between the computer and the stereoscopic image-synthesizer (Synthesizer) to identify the bar code combination through different colors. (Representing different functional requirements) 'Can tell the stereoscopic synthesizer to synthesize the images of the separate images into a stereoscopic image, and at the same time erect or attach a grating plate to the window of the display (Lenticular Lens) ), allowing the user to view the stereoscopic image on the flat panel display without wearing stereo glasses. [Embodiment] The detailed description and technical description of the present invention will now be described as follows: Please refer to "Fig. 1", which is a multi-view stereoscopic image display processing method of the present invention (Multiple Views of Stereoscopic Image Display) Scheme) Schematic diagram of the process. As shown in the figure, the multi-view stereoscopic image display processing method of the present invention includes: a multi-view (Multiple)

Views) image 1, image format separation processing (viewArrangement) 2, a color recognition barcode (Color Codes) frame (Frame) 3 and a stereoscopic image synthesizer (Synthesizer) 4. After obtaining multiple views (Multiple Views) images, the multiple views (Multiple Views) images are image-separated (as shown in Figures 2 - 1 to 2 - 3), and then the pre-planned image represents the image. The separated color recognition barcode (col〇rC〇des) image is sent to inform the stereoscopic image synthesizer (Synthesizer) what kind of multi-view image is to be arranged, then the stereo image synthesis is performed, and the separated image is synthesized. The display of the stereoscopic image is at the same time that the window of the flat display device is erected or attached with the line-line #6, so that the manufacturer can view the stereoscopic image on the flat display 5 without having to call the mirror. ^ First, in order to achieve the above-mentioned stereoscopic image display, the first step is to obtain a multi-view image, which is to use the image capturing device (such as a digital camera or a computer simulated by a computer) in the same-horizontal plane. According to the straight (or arc) path, the image of the image capture can be displayed or all the same - the multi-view image taken by the object in a unique way (following the spider 7 K78225

Views); After the above steps are completed, the image format separation process 2 is performed, and the multi-view images captured as described above are also separated (View Arrangement, such as the second - i~2 - 3 map (FlatPanel Display) ^^(Multiple Views) Views are zoomed out and evenly arranged in the same face. Each image size is: (flat display horizontal resolution / χ#) χ (flat display vertical resolution For example, in the level of 1G24x768pixels, you can arrange the image of the image as 4仃X2 on average, so each image size is 256x384pixds; similarly, you can put 9 (or 10) The average image of the rural image is _ 3 lines _ (or 5 lines of lake), 0 each of which has a size of 341x256 pixels (or 204 x 384 pixels), and each of the separated and reduced images is taken at different angles (ie, each is reduced) The images are independent of each other;; after the above-mentioned shirt image separation processing 2 is completed, then the color identification barcode (c〇1〇rc〇des) 3 which is arranged in advance to represent the multi-view image is displayed in one Independent page (Frame) 'When the stereoscopic image synthesizer 4 receives the color identification barcode (ColorCodes) 3 1 (as shown in Fig. 3), it can recognize that the multi-view image sent is separated into several images and arranged. Why, at this time, the stereoscopic image synthesizer will determine the resolution of the surface after the decoding of the color code barcode (Color Codes) 3 1 , the line of the multi-view image separation (c〇hmm), the column (R 〇w), calculate the number of images (ViewNumber) and the size of each view image (ViewSize), and then divide the horizontal face into (horizontal resolution / line) according to the resolution of the face (as shown in Figure 5) Block (B1〇ck), the vertical plane is divided into (vertical resolution/column) blocks (9) ock), and each block (9) 〇 ck) corresponds to the pixel of each image separated by the multi-view image. (Pixel or Pictlire Element); that is, the basic constituent unit of the naked-eye stereoscopic image of the present invention is the block, and the basic constituent unit of the block is composed of pixels of each view image. Since each pixel (Pixel) of a flat panel display (such as a liquid crystal screen) is composed of red (R), green (G), and blue (9) three sub-pixels (as shown in Fig. 6). At the same time, there is a small gap between the three light elements (Sub-Pixel) of r, g, and B. If the light plate (Lenticuiar Lens) is simply added, it will be seriously caused by this black gap (BlackMatrix). The diffraction interference (Morie )' so we can rotate or tilt the Lenticular Lens to a 0 angle of about 8 Γ 278225 9.4623 (since each Sub-Pixel has a vertical length that is three times the horizontal length,

Sub-Pixel is enough to cover this black gap (BlackMatrix) 7, so according to the trigonometric relationship tan6» 1/6 ′ is 1/6=9.4623. , please refer to Figure 7), you can cover this black gap and solve the problem of optical interference. In order to match the angle of the grating plate (LemicuiarLens), the pixels (pixei) of each view image in the block must also be relatively adjusted to correspond to the inclination of the grating piece, and must be R, G. Time element

The stereoscopic image synthesizer (gyntheSiZer) of the present invention is a hardware processor V (Ware Processor), and may also be a computer software simulator (Software Simulator), whose main function is to provide a solution to this, R, G, The B-light element (Sub_pixd) is a stereoscopic image of the basic arrangement unit. The method of synthesizing the method is as follows: 下列8〇ι^^): Combining the multi-images composed of the separated row-riding--a stereo image The facet output (as shown in Fig. 4); the aforementioned color identification bar code (c〇1〇rC〇des) 3 i can be at least one of the two colors; $ 1278225 calculus (A丨gorithm) description: ____ Each view image is stored in the corresponding block (Block) (Block-X) X (Block)^fc). Re-discharge the image stored in each block (Block) as the light element Sut>Pixel) in its corresponding position (must match the grating specifications, LCD Dot Pitch, View number, and i different). Each View pixel stored in each block (a total of (a〇ck_4 X (Block_y) points) will be remapped to a different small Map Block (the size is (N__x) x (N__y point)) Total (Block)〇x Map Block. N (View number) 8'9,10,12 N = 8 N = 9 N = :10 N—x (horizontal block number) 4 3 5 N_y (vertical block Number) 2 3 2 Res-x (horizontal resolution) 1024, 1280, 1600, · Res_y (vertical resolution) 768, 1024, 1200, ·· Block-x (each block width) Res-X DIV N-X Block_y (each block height ^ Res_y DIV N_y Block_Start (starting station Line_Start of each block (starting bit of each block row Source(x, y) (View pixel to be re-discharged within the block) Destination(X, Y) (The View pixel to be re-discharged Source(x, y), which is to be re-discharged in the address of the small Map Block where the Destination is located.) Processing Algorithm: For (n = 0 to N1) // Process the View images stored in each block in order. // Find the starting address of each block (Block Start) Block-Start = (n DIV N_y) MUL BTock^y MUL Res"c ADD (n MOD N_y) MUL Block_x LJDestination (X,Y) = 〇For(y=1 to Block__y) // Each Block , in sequence, a Block Row - Block Row processing. " Find the starting address of each Block Row (Linejtart). Line One Start = Block - Start ADD (y SUB 1) MUL Res - x

10 1^78225

Destination(X,Y) = L_Destination(X,Y) ___ For(x=1 to Block_x) { One //Each Block Row, in sequence one Pixel - one Pixel processing. // Find the View pixel (Source(x, y)) that is being re-discharged. Source(x,y) = Line_Start ADD (X SUB1) //Re-discharge R, G, and B of this point Source(x, y) to the corresponding position. //R of this point should be relocated to R of the address (X1, Y1) where Destination is located. //G at this point should be relocated to G° of the address where the Destination is located (X2, 丫2) // B at this point should be relocated to B of the address (X3, Y3) where Destination is located. DestinationRGB(X1,Y1,R) = Source(x,y),R DestinationRGB(X2, Y2, G) = Source(x,y). G DestinationRGB(X3, Y3, B) = Source(x,y). B // Horizontally move the small M叩Block where the Destination is located, and move to the next // Small Map Block. Destination(X, Y) = Destination(X, Y) ADD N_x } A // Vertically move the small Map Block where the Destination is located (N_y x N_y point), // Move directly to the small Map Block of the next column. L-Destination(X,Y) = L-Destinatlon(X,Y) ADD I(ISLy MUL Res-x) Supplement: _____ DestinationRGB(X, Y,R) is the location of the location ^Y) R Sub - The location of Pixel. DestinationRGB(X, Y, G) is the location of the G Sub -Pixel at the location of Destination%, Y). DestinationRGB(X,Υ,B) is the location of B Sub -Pixe丨 at the point where Destination(X,Y) is located. DestinationRGB(X,Y) is the reference image point Source(x,y), corresponding to the address of the small Map Block where the Destination is located. And R, G, B really want to map the address placed: DestinationRGB (X1, Y1, R), DestinationRGB (X1, Y1, G), DestinationRGB (X1, Y1, B) is the address corresponding to this small Map Block DestinationRGB(X,Y) plus an offset (Offset) is determined by Fn(n, Lx, rgb). DestinationRGB(X1, Y1, R) = DestinationRGB(X, Y) + Fn(n, Lx, R) DestinationRGB(X2, Y2, G) = DestinationRGB(X, Y) + Fn(n, U, G)

11 1278225 where Fn(n, Lx, rgb) is a Hash Map Table used to calculate the displacement (〇ffset) of a corresponding small Map Block address. Parameters: π - the number of views, n = 8, 9, 10, 12 This Fn has a different Map Table Lx = y MOD N_x (different Destination Row (y) will have different Map Value) rgb = R or G or B ^ N=8 n = 0 〜7 Lx = 0 〜3 Rgb = R, G, B Fn(0, 0, R) = 1 Fn(0, 0, G) = 5 Fn(0, 0, B) = 9 Fn(1,0, R) = 2 + Res-x * 3 Fn(1,0, G) = 6+ Res-x * 3 Fn(1,0, B) = 10+Res_x*3 Fn(2, 0, R) = 2 Fn(2, 0, B)=10 Fn(2, 0, G) = 6 Fn(3,0,R) = 3+ReS-X*3 Fn(3, 0, G) = 7+ Res-x * 3 Fn(3, 0,B)=11+Res—x*3 Fn(4, 0, R) = 3 Fn(4, 0, G) = 7 Fn( 4, 0, B) = 11 Fn(5, 0, R) = 4+ Res—x * 3 Fn(5, 0, G) = 8+ Res-x * 3 Fn(5, 〇, B) = 12 + Res-x * 3 Fn(6, 0, R) = 4 Fn(6, 0, G) = 8 Fn(6, 〇, B) = 12 Fn(7, 〇, R) = 4+ Res-x * 3 Fn(7, 〇, G) = 8+ Res-x * 3 Fn(7, 〇, B)= 12+ Res-x*3 Stereo Image Synthesizer Processing Algorithm 12 1278225 / by It can be seen that the present invention is a multi-view image taken by a part or more of the image capturing device, and the multi-view image is separated and arranged (for example, the second to the second 3), after the image separation processing (ViewArrangement), send a color identification stop code (Color Codes) face (Frame) 3 (as shown in Figure 3), in the stereo image synthesizer (Symhesizer) 4 After receiving the color identification barcode (c〇1〇rC〇des), it can be judged that the multi-view image is separated into several images, and how the separated images are arranged, and then R is used. , G, B light element (Sub-Pixd) stereo image synthesis method, the separated picture is synthesized into a stereo image, the stereo image can be displayed on the flat display 5, and then through the crying - 5 window Tilt 9·4623. With the help of the grating plate and the coffee café Lens 6, the user can not wear the stereo glasses T, and can view the vertical image on the plane age ϋ 5 . Or, it is that the computer processing work is fast enough, and it is enough for the instant (four) D (four) operation of the φ 下 ' can be omitted according to actual needs - image format separation processing (ViewArrang_nt) 2, a color recognition barcode (ColorCodes) frame (Frame) 3 Processing steps, and directly send the multi-view images (MuMple Views) to the stereoscopic image synthesizer simulated by the computer software. She informs the multi-view image number (ViewNumber) and the horizontal level and vertical display resolution (Resolution) by (4) 4 ' ) 'The stereo image synthesis process can be performed, and the result can also be on the flat display 5, and the stereoscopic image can be viewed without wearing the stereo glasses. The above is only the preferred embodiment of the invention and is not intended to limit the scope of the invention. That is, the equivalent changes and modifications made by the scope of the patent application of the present invention are covered by the scope of the invention. ▲ 13 1278225 [Simplified description of the drawings] Fig. 1 is a schematic flow chart of the method for processing stereoscopic image display according to the present invention. Figures 2-1 to 2-3 are three schematic views of the multi-view image separation arrangement of the present invention.苐3图' is a schematic diagram of the color identification barcode (c〇i〇rCodes) of the present invention. Fig. 4 is a schematic diagram of a color decoder (c〇1〇r Decoder) included in the stereoscopic image synthesizer of the present invention. Fig. 5 is a block diagram of a block (B1〇ck) included in the flat panel display (FlatPanelDisplay) of the present invention. The figure is a schematic diagram of the structure of the block (Bk) ck of the present invention.

Fig. 7 is a schematic view showing the rotation of a grating plate (Lenticuiar Lens) of the present invention at a 0 angle. [Main component symbol description] ...........Multiple Views image capture 2 ...Image Arrangement 3 .....Color recognition barcode ( Color Codes) Frames 4 ..........Synthesizer 31..........Color Codes 5 ..... ...Flat Panel Display

6 ... tilt 9.4623. Lenticular Lens .............Black Matrix 14

Claims (1)

Ιυ t 1278225 : 2, \ 4 X. Patent application: 1. Multi-view view of Stereoscopic Image display processing method is applicable to a display with a grating plate, which includes the following steps: a) First, multi-view images (MultipleViews); b), the above-mentioned multi-view image for image format separation (view Arrangement), according to the flat panel display (Flat Panel Display) specifications and the number of multi-view images can be separated into 8 Grid, 9 or 1 〇, ..., etc.; c), wait, after the above image format separation process is completed, a color code barcode (Frame) is sent out, which is received by the synthetic image processor. This color recognition bar code (Color C〇des) can be used to identify the separated image. The image is separated from several images, and the image is separated by the knife. The Synthesizer synthesizes the above-mentioned separate images by using the R, G, B light element (Sub-Pixel) stereo image synthesis algorithm based on the identification bar code. Body image output. 2. The multi-view stereoscopic image display processing method according to claim 1, wherein the multi-view image is a one or more image capturing devices (such as a digital camera or a computer-simulated camera) In the same horizontal plane, according to the straight line (or arc) path, the lens of the image capturing device can be arranged in parallel or aligned with the same object to shoot multiple views (Multiple Views) at different angles. In the multi-view stereoscopic image display processing method described in the above item, the separated image plane can be separated into a plurality of image arrangement screens, and each of the separated image frames is photographed at different angles. 4. The multi-view stereoscopic image display processing method according to claim 1, wherein the color identification barcode (ColorCodes) can be displayed in at least one of one or more modes. 5. The multi-view stereoscopic image display processing method according to claim 1, wherein the stereoscopic image synthesizer replaces the conventional pixel with the three light elements (Sub-Pixel) of R, G, and B (Pixel). As a unit of stereoscopic image synthesis, and according to the image synthesis processing algorithm CProeessing Algorithm) for stereoscopic image synthesis. The multi-view stereoscopic image display processing method according to claim 1, wherein the display device can be a general flat panel display (Fiat panei display such as LCD monitor). The method of processing a multi-view stereoscopic image display according to claim 6, wherein the window of the flat display is erected or attached with a grating plate having a vertical inclination of about 9.4623° (degrees) ( Lenticular Lens) 〇1278225 VII. Designated representative map: (1) The representative representative of the case is: (^). (2) A brief description of the component symbols of this representative diagram: 2 · 3 · 4 · 5 · 6 · •• Multiple Views image capture • View Arrangement • Color Codes昼Frame ••Synthesizer •••Flat Panel Display • ·· Tilt 9.4623 Lenticular Lens If there are chemical formulas in the eight cases, please reveal the best indication of the invention. Chemical formula