WO2017124586A1 - 一种裸眼3d显示方法及系统 - Google Patents

一种裸眼3d显示方法及系统 Download PDF

Info

Publication number
WO2017124586A1
WO2017124586A1 PCT/CN2016/073084 CN2016073084W WO2017124586A1 WO 2017124586 A1 WO2017124586 A1 WO 2017124586A1 CN 2016073084 W CN2016073084 W CN 2016073084W WO 2017124586 A1 WO2017124586 A1 WO 2017124586A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
signal
brightness
display
depth map
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2016/073084
Other languages
English (en)
French (fr)
Inventor
徐遥令
侯志龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Skyworth RGB Electronics Co Ltd
Original Assignee
Shenzhen Skyworth RGB Electronics Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Skyworth RGB Electronics Co Ltd filed Critical Shenzhen Skyworth RGB Electronics Co Ltd
Priority to AU2016304612A priority Critical patent/AU2016304612B2/en
Priority to US15/507,494 priority patent/US10326974B2/en
Publication of WO2017124586A1 publication Critical patent/WO2017124586A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
    • H04N13/106—Processing image signals
    • H04N13/128—Adjusting depth or disparity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L13/00—Details of the apparatus or circuits covered by groups H04L15/00 or H04L17/00
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
    • H04N13/106—Processing image signals
    • H04N13/111—Transformation of image signals corresponding to virtual viewpoints, e.g. spatial image interpolation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
    • H04N13/106—Processing image signals
    • H04N13/15—Processing image signals for colour aspects of image signals
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
    • H04N13/106—Processing image signals
    • H04N13/161—Encoding, multiplexing or demultiplexing different image signal components
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/30—Image reproducers
    • H04N13/302—Image reproducers for viewing without the aid of special glasses, i.e. using autostereoscopic displays
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/597—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding specially adapted for multi-view video sequence encoding

Definitions

  • the present invention relates to the field of naked eye 3D display, and more particularly to a naked eye 3D display method and system.
  • a 3D (three-dimensional) image is usually composed of two images, which can achieve a 3D effect in a human eye corresponding to a specific viewing angle, forming a three-dimensional effect, which makes the eyes look like real. .
  • an image corresponding to the new viewing angle can be generated, which is the key to the naked-eye 3D TV.
  • the naked-eye 3D display technology allows two left and right eyes to see two disparate and different images from the display screen without any tools. Reflecting them into the brain, people will have a three-dimensional effect. It also utilizes the parallax principle of the human eye to achieve a three-dimensional visual effect by sending different images to the viewer's left and right eyes. Since the observer of the naked-eye 3D TV can perform the 3D display experience without wearing glasses, it meets the market demand of the 3D display, and has a large market and business opportunities.
  • 3D signals output by 3D source devices are generally left and right (L/R) 2 viewpoints
  • naked-eye 3D TVs require more viewpoints for 3D experience over a wide range, so it is necessary to convert 2 viewpoints into multiple viewpoints, and Viewpoint conversion requires additional TV resources, and viewpoint conversion reduces viewing clarity Wait.
  • the technical problem to be solved by the present invention is to provide a naked-eye 3D display method and system, which aims to solve the problem that the existing 3D viewpoint conversion needs to occupy additional television resources and reduce the definition.
  • a naked eye 3D display method comprising the following steps:
  • S4 Perform frame interpolation and image calibration on the composite image, and send the image to a naked-eye 3D display to implement naked-eye 3D display.
  • step S2 specifically includes:
  • step S21 specifically includes:
  • S211 Perform signal separation on the original 3D video signal, and convert the original 3D image signal into a left image signal, a right image signal, and an L/R synchronization signal corresponding to the left image signal and the right image;
  • S215 regularly arrange pixels in the L image and the R image of the RGB color space, and regularly sew the R, G, and B components in the pixel into an LVDS signal format, and insert the four pixels in the earliest transmission of the L image.
  • the L flag inserts the R flag into the four pixels of the earliest transmission of the R picture, and alternately outputs at a frame rate of 2f1.
  • step S212 specifically includes:
  • S2122 analyzing and categorizing the brightness Y in the image signal of the YUV color space, and dividing into a low brightness area, a medium brightness area, and a high brightness area according to the brightness value; the brightness of the low brightness area is less than the first critical value, The brightness of the medium brightness region is between the first critical value and the second critical value, the brightness of the high brightness region is greater than the second critical value; the first critical value is less than the second critical value;
  • S2124 uses hierarchical interpolation to improve the resolution, and uses different interpolation algorithms for different luminance regions to convert the image resolution into a fixed resolution.
  • the converted L image and R image pixels are M*N arrays.
  • step S3 specifically includes:
  • S31 performing de-weaving and image separation on the image signal of the LVDS signal format, separating the L image sequence and the R image sequence; acquiring depth information of the object object in the L image sequence and the R image sequence, and estimating the depth of each pixel point.
  • Information forming a continuous smooth depth map; estimating a plurality of L viewpoint images according to the L image and the corresponding depth map, estimating a plurality of R viewpoint images according to the R image and the corresponding depth map; and comparing the plurality of L viewpoint images and the plurality of The R view image is subjected to interleaving processing to form a composite image including a plurality of viewpoints; and the composite image is format-woven to obtain a corresponding VBO signal format image signal.
  • step S31 specifically includes:
  • step S4 includes:
  • the calibrated composite image is woven into a corresponding VBO signal format image signal, and transmitted to the naked-eye 3D display at a frequency of Tf2 to realize naked-eye 3D display.
  • a naked eye 3D display system comprising:
  • a decoding module configured to decode the 3D signal to obtain an original 3D video signal
  • a first 3D video processing module configured to perform signal separation, image enhancement, and classification into the original 3D video signal to obtain a corresponding L image and an R image;
  • a second 3D video processing module configured to acquire depth information of an object object in the L image and the R image, and estimate depth information of each pixel to form a corresponding depth map; and estimate multiple according to the L image and the corresponding depth map
  • An L view image estimating a plurality of R view images according to the R image and the corresponding depth map; performing interleaving processing on the plurality of L view images and the plurality of R view images to form a composite image including the plurality of views;
  • the third 3D video processing module is configured to perform frame interpolation and image calibration on the composite image, and send the image to a naked-eye 3D display to implement naked-eye 3D display.
  • the naked eye 3D display system wherein the first 3D video processing module comprises:
  • a first 3D video processing unit configured to perform signal separation on the original 3D video signal to obtain a corresponding L image signal and an R image signal; perform image lifting and classification on the L image signal and the R image signal to obtain a corresponding image L image sequence and R image Sequence; format encoding the L image sequence and the R image sequence to obtain a corresponding LVDS signal format image signal.
  • the naked eye 3D display system wherein the first 3D video processing unit comprises:
  • a signal separation unit configured to perform signal separation on the original 3D video signal, and convert the original 3D image signal into a left image signal, a right image signal, and an L/R synchronization signal corresponding to the left image signal and the right image;
  • the color conversion and image lifting unit is configured to convert the color space image of the L image and the R image into a YUV color space, and perform image enhancement and brightness conversion on the L image and the R image of the YUV color space by using luminance partition scaling and hierarchical interpolation.
  • the classification inserting unit is configured to classify the L image and the R image with the pixel M*N under the control of the L/R synchronization signal, sequentially cache the L image together to form an L image set, and cache the R image together.
  • R image set respectively inserting L image set and R image set into frame, forming L image set with frame frequency f1 and R image set with frame frequency f1; sequentially outputting images of L image set and R image set, and Output corresponding to the R identification signal and the L identification signal; wherein, f1 is a positive integer;
  • An RGB image conversion unit for performing RGB image conversion on the L image and the R image, and converting the L image and the R image of the YUV color space into the L image and the R image of the RGB color space;
  • a first woven output unit for regularly arranging pixels in the L image and the R image of the RGB color space, and arranging the R, G, and B components in the pixel into LVDS according to rules
  • the signal format, and the L flag is inserted into the four pixels of the earliest transmission of the L picture, and the R flag is inserted into the four pixels of the earliest transmission of the R picture, and alternately output at a frame rate of 2f1.
  • the naked eye 3D display system wherein the color conversion and image lifting unit comprises:
  • a color conversion unit for converting a color space image of the L image and the R image into a YUV color space, and converting pixels of the L image and the R image into pixels of YUV 4:2:2;
  • the brightness classification unit is configured to analyze and classify the brightness Y in the image signal of the YUV color space, and divide the brightness value into a low brightness area, a medium brightness area and a high brightness area according to the brightness value; the brightness of the low brightness area is less than the first critical a value, the brightness of the medium brightness region is between a first threshold value and a second threshold value, the brightness of the high brightness area is greater than a second threshold value; the first threshold value is less than a second threshold value;
  • the partitioning and scaling unit is configured to use brightness partitioning to enhance the contrast, to perform brightness compression on the image in the low-light area, to enhance the brightness of the image in the medium-light area, and to maintain the brightness of the high-brightness area unchanged;
  • the hierarchical interpolation unit is used to increase the resolution by using hierarchical interpolation, and different interpolation algorithms are used for different brightness regions to convert the image resolution into a fixed resolution, and the pixels of the converted L image and the R image are M*N arrays. .
  • the naked eye 3D display system wherein the second 3D video processing module comprises:
  • a second 3D video processing unit configured to perform de-weaving and image separation on the LVDS signal format image signal, and separate the L image sequence and the R image sequence; and acquire the L image Forming depth information of the object object in the sequence and the R image sequence and estimating the depth information of each pixel to form a continuous smooth depth map; estimating a plurality of L viewpoint images according to the L image and the corresponding depth map, according to the R image and Corresponding depth maps are used to estimate a plurality of R view images; a plurality of L view images and a plurality of R view images are interleaved to form a composite image including a plurality of viewpoints; and the composite image is format-woven to obtain a corresponding VBO signal. Format image signal.
  • the naked eye 3D display system wherein the second 3D video processing unit comprises:
  • a de-weaving and image separating unit for converting an LVDS signal format image signal into an RGB image signal, separating the L image and the R image;
  • a depth map generating unit for acquiring similar object objects in the L image and the R image, and calculating a parallax of the similar object object in the L image and the R image according to the horizontal displacement of the similar object object in the L image and the R image; And processing the color, texture and light and shadow information of the R image object, estimating the depth information of each pixel, and forming a continuous smooth depth map;
  • a multi-viewpoint generating unit configured to estimate a plurality of L view images according to the L image and the depth map by using an interpolation view or an epitaxial view, and estimate a plurality of R view images according to the R image and the depth map;
  • a view synthesis unit configured to perform a process of interleaving a plurality of L view images and a plurality of R view images according to a physical pixel arrangement of the naked eye 3D display, and synthesizing a composite image including the plurality of view information
  • a second weaving output unit for performing a grid on a composite image including a plurality of viewpoint information Weaving, weaving into a corresponding VBO signal format image signal, output at a speed of f2, the f2 being a positive integer.
  • the naked eye 3D display system wherein the third 3D video processing module comprises:
  • a de-weaving unit for de-weaving the image signal of the VBO signal format, and converting the image signal of the VBO signal format into a composite image of the RGB signal format;
  • the frame interpolation frequency multiplying unit is configured to perform frame interpolation and frequency multiplication processing on the composite image, and repeat one frame image to become the same T frame image, thereby obtaining a composite image of frequency Tf2, wherein T is greater than 1;
  • An image calibration unit for performing image calibration on the composite image according to the electrical characteristics and display effect requirements of the naked eye 3D display
  • the woven output and display unit is configured to woven the calibrated composite image into a corresponding VBO signal format image signal, and transmit the signal to the naked-eye 3D display at a frequency of Tf2 to realize naked-eye 3D display.
  • the naked eye 3D display method and system provided by the invention effectively solve the problem that the existing 3D viewpoint conversion needs to occupy additional television resources and reduce the definition, and the original 3D video signal is obtained by decoding the 3D signal;
  • the original 3D video signal is used for signal separation, image enhancement and classification into frame, and the corresponding L image and R image are obtained;
  • the depth information of the object object in the L image and the R image is obtained, and the depth information of each pixel is estimated to form a corresponding a depth map; estimating a plurality of L viewpoint images according to the L image and the corresponding depth map, estimating a plurality of R viewpoint images according to the R image and the corresponding depth map; and interleaving the plurality of L viewpoint images and the plurality of R viewpoint images Forming a package after processing a composite image with multiple viewpoints; frame interpolation and image calibration of the composite image, and transmission to a naked-eye 3D display to achieve naked-eye 3D display; efficient realization of naked-eye 3D viewpoint conversion, less resource occupation,
  • FIG. 1 is a flow chart of a first embodiment of a naked eye 3D display method provided by the present invention.
  • FIG. 2 is a flow chart of a method for a second embodiment of a naked eye 3D display method according to the present invention.
  • FIG. 3 is a schematic flowchart of a method of step S210 in a second embodiment of a naked eye 3D display method according to the present invention.
  • FIG. 4 is a schematic flow chart of the method of step S310 in the second embodiment of the naked eye 3D display method provided by the present invention.
  • FIG. 5 is a schematic flowchart of a method of step S410 in a second embodiment of a naked eye 3D display method according to the present invention.
  • FIG. 6 is a structural block diagram of a preferred embodiment of a naked-eye 3D display system provided by the present invention.
  • FIG. 7 is a schematic diagram of a 3D video signal in an application embodiment of a naked-eye 3D display system provided by the present invention.
  • FIG. 8 is a schematic diagram of an LVDS signal format in an application embodiment of a naked eye 3D display system provided by the present invention.
  • FIG. 9 is a schematic diagram of a naked-eye 3D display in an application embodiment of a naked-eye 3D display system provided by the present invention.
  • the present invention provides a naked eye 3D display method and system.
  • the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
  • FIG. 1 is a flowchart of a first embodiment of a naked-eye 3D display method according to the present invention. As shown in the figure, the method includes the following steps:
  • Step S100 decoding the 3D signal to obtain an original 3D video signal
  • Step S200 performing signal separation, image enhancement, and classification into the original 3D video signal to obtain a corresponding L image and an R image;
  • Step S300 Obtain depth information of the object object in the L image and the R image, and estimate depth information of each pixel to form a corresponding depth map; and estimate a plurality of L viewpoint images according to the L image and the corresponding depth map, according to R Estimating a plurality of R view images by the image and the corresponding depth map; performing interleaving processing on the plurality of L view images and the plurality of R view images to form a composite image including the plurality of viewpoints;
  • Step S400 Perform frame interpolation and image calibration on the composite image, and send the image to a naked-eye 3D display to implement naked-eye 3D display.
  • step S100 the 3D signal is decoded to obtain an original 3D video signal.
  • the naked eye 3D can be applied to a terminal having a display screen such as a television, a mobile phone, or a tablet computer.
  • the most widely used ones are naked-eye 3D televisions, and the invention is described below by taking naked-eye 3D television as an example.
  • 3D signal is 3D TV signal, interface circuit of naked eye 3D TV
  • the 3D television signal outputted by the external 3D signal source device is received, subjected to amplitude limitation, format recognition, and the like, and the processed 3D television signal is output to the signal decoding portion; the decoding portion decodes the 3D television signal to obtain the original 3D video signal.
  • step S200 signal separation, image enhancement, and classification of the original 3D video signal are performed to obtain a corresponding L image and R image.
  • FIG. 2 is a flowchart of a method for the second embodiment of the naked eye 3D display method according to the present invention.
  • the step S200 includes: S210, The original 3D video signal is separated to obtain a corresponding L image signal and an R image signal; the L image signal and the R image signal are image-enhanced and classified into a frame to obtain a corresponding L image sequence and an R image sequence; The sequence and the R image sequence are formatted to obtain a corresponding LVDS signal format image signal.
  • signal separation is performed on the original 3D video signal to obtain an original L (left eye) image signal and an R (right eye) image signal, and an L/R synchronization signal; color conversion is performed on the L and R image signals, and luminance partitioning is performed.
  • Scaling and grading interpolation for image enhancement conversion to L and R images with pixels as M*N arrays; classification and illustration of L and R images with pixels M*N, converting them into L image sequences of frequency f1 And an R image sequence having a frequency of f1; performing RGB image conversion on the L image and the R image in the image sequence, the pixels in the image are composed of R, G, and B, and further pattern-weaving the L image sequence and the R image sequence,
  • the L image and its L mark are woven into an LVDS (Low Voltage Differential Signaling) signal format, and the R image and its R mark are also woven into an LVDS signal format, and alternately output at a frame rate of 2f1.
  • LVDS Low Voltage Differential Signaling
  • M, N and f1 are positive integers, and the specific values can be set.
  • L map The image represents the left image
  • the R image represents the right image
  • the L image, the R image, and the L/R logo are commonly used in the field of 3D display technology.
  • the step S210 specifically includes:
  • S211 Perform signal separation on the original 3D video signal, and convert the original 3D image signal into a left image signal, a right image signal, and an L/R synchronization signal corresponding to the left image signal and the right image;
  • S215 regularly arrange pixels in the L image and the R image of the RGB color space, and regularly sew the R, G, and B components in the pixel into an LVDS signal format, and insert the four pixels in the earliest transmission of the L image.
  • the L flag inserts the R flag into the four pixels of the earliest transmission of the R picture, and alternately outputs at a frame rate of 2f1.
  • FIG. 3 is a schematic flowchart of a method of step S210 in a second embodiment of a naked-eye 3D display method according to the present invention.
  • the original 3D video signal is separated.
  • the original 3D image signal is converted into a left (L) image signal and a right (R) image signal, and an L/R sync signal corresponding to the L image signal and the R image.
  • the L image and the R image of the YUV color space are image-enhanced by luminance partition scaling and hierarchical interpolation, and converted into L and R images whose pixels are M*N arrays; wherein M and N are positive integers.
  • the L image and the R image are classified, and the L image is sequentially cached to form an L image set, and the R image is cached together to form an R image set; respectively, the L image set and the R image are respectively
  • the set is interpolated to form an L image set having a frame rate of f1 and an R image set having a frame rate of f1.
  • the images in the L image set and the R image set are sequentially output, and the corresponding R and L identification signals are output.
  • f1 is a frequency value and is a positive integer.
  • the L image and the R image are further subjected to RGB image conversion, and the image of the YUV color space is converted into an image of the RGB color space, and the pixels in the RGB image are composed of R, G, and B.
  • step S300 weave the output and arrange the pixels in the L image and the R image regularly. And arranging the R, G, and B components in the pixel into an LVDS (Low Voltage Differential Signal) transmission format, and inserting the L flag in the earliest transmitted four pixels of the L image, and the four pixels of the earliest transmission in the R image. The R flag is inserted in the middle, and finally output at a frame rate of 2f1. As shown in FIG. 3, the output LVDS image signal can be recorded as the first 3D video signal. The processing as described in step S300 is then performed on the first 3D video signal.
  • LVDS Low Voltage Differential Signal
  • the step S212 specifically includes:
  • S2122 analyzing and categorizing the brightness Y in the image signal of the YUV color space, and dividing into a low brightness area, a medium brightness area, and a high brightness area according to the brightness value; the brightness of the low brightness area is less than the first critical value, The brightness of the medium brightness region is between the first critical value and the second critical value, the brightness of the high brightness region is greater than the second critical value; the first critical value is less than the second critical value;
  • S2124 uses hierarchical interpolation to improve the resolution, and uses different interpolation algorithms for different luminance regions to convert the image resolution into a fixed resolution.
  • the converted L image and R image pixels are M*N arrays.
  • image enhancement of the L image and the R image of the YUV color space includes the following steps:
  • the brightness Y of the image signal in the YUV color space is analyzed and classified into: A low luminance region (luminance is smaller than the first critical value Yd), a high luminance region (brightness is greater than the second threshold value Yh), and a medium luminance region (the luminance is between the first critical value Yd and the second critical value Yh).
  • the first threshold value Yd is smaller than the second threshold value Yh, and may be specifically set according to actual needs.
  • the brightness partitioning is used to increase the contrast: the brightness of the image in the low-light area is compressed, that is, the brightness Pd of the pixel in the low-light area (Pd is the brightness value of the pixel in the low-light area) becomes Pd*Cd, and Cd is a variable. And Cd is not greater than 1; brightness enhancement of the image of the medium-bright area, that is, the brightness Pz of the pixel in the middle bright area (Pz is the brightness value of the pixel in the bright area) becomes Pz*Cz, Cz is a variable, and Cz Not less than 1 and Pz*Cz is not greater than the critical value Yh of the highlighted area. The brightness of the high-brightness area remains unchanged.
  • the contrast is improved, and the brightness of the high-brightness area is unchanged, and the image is not distorted.
  • the hierarchical interpolation is used to improve the resolution, that is, different interpolation algorithms are used for different luminance regions, and the image resolution is converted into a fixed resolution, and the pixels of the converted L image and the R image are M*N arrays. Specifically:
  • the low-brightness area and the high-brightness area are interpolated by a four-point averaging algorithm, that is, the four adjacent pixels around the pixel Px to be interpolated are Pa(Ya, Ua, Va), Pb(Yb, Ub, Vb), respectively.
  • Y, U, V, etc. respectively represent the component values of the four pixels in the YUV color space, which are known in the prior art and will be easily understood by those skilled in the art and will not be described again.
  • the medium luminance region is interpolated by a close association algorithm, that is, four known pixels adjacent to the pixel Px to be interpolated are Pa(Ya, Ua, Va), Pb (Yb, Ub, Vb), Pm (Ym). , Um, Vm), Pn (Yn, Un, Vn), the inserted pixel Px is (Yx, Ux, Vx).
  • Yx, Ux, Vx calculation process is: taking Yx as an example,
  • step S300 acquiring depth information of the object object in the L image and the R image and estimating depth information of each pixel point to form a corresponding depth map; and estimating a plurality of L viewpoint images according to the L image and the corresponding depth map, A plurality of R view images are estimated from the R image and the corresponding depth map; and the plurality of L view images and the plurality of R view images are interleaved to form a composite image including a plurality of viewpoints.
  • the step S300 specifically includes: S310, performing de-weaving and image separation on the LVDS signal format image signal, separating the L image sequence and the R image sequence; acquiring the L image sequence and the R image. Depth information of the object object in the sequence and estimating the depth information of each pixel to form a continuous smooth depth map; The L image and the corresponding depth map are used to estimate a plurality of L view images, and a plurality of R view images are estimated according to the R image and the corresponding depth map; and the plurality of L view images and the plurality of R view images are interlaced to form a plurality of images. a composite image of the viewpoints; the composite image is format-woven to obtain a corresponding VBO signal format image signal.
  • the LVDS signal format image signal (that is, the first 3D video signal) is subjected to de-weaving and L/R image separation processing to separate the L image and the R image; and processing the L image and the R image, Obtaining depth information of the object object in the image and estimating the depth information of each pixel to form a continuous smooth depth map; estimating the plurality of L viewpoint images according to the L image and the depth map by using the interpolation viewpoint or the epitaxial viewpoint And estimating a plurality of R view images according to the R image and the depth map to form a plurality of view images; and interleaving the plurality of view images to form the plurality of view points according to the physical pixel arrangement of the ultra high definition naked eye 3D screen
  • the composite image has a pixel number of H*V, where the value of H/M is an integer and the value of V/N is an integer. And further synthesizing the composite image including the multi-viewpoint information, weaving it into a VBO (V-by-One, which
  • the step S310 specifically includes:
  • FIG. 4 is a schematic flowchart of a method of step S310 in a second embodiment of a naked eye 3D display method according to the present invention.
  • the de-weaving and L/R image separation are first performed: receiving the LVDS signal format image signal, converting the signal of the LVDS transmission format into an RGB image signal, and obtaining the L mark or the R mark in the process, The L mark separates the L image and separates the R image by the R mark.
  • perform depth map generation acquire similar object objects in the L image and the R image, and calculate the depth (or called parallax) of the similar object in the L image and the R image according to the horizontal displacement of the similar object in the L image and the R image. Then, according to the color, texture, light and shadow information of the L image and the R image object, the depth information of each pixel is estimated, and a continuous smooth depth map is further formed.
  • multi-view generation can be done by interpolation of viewpoints, or The way to look at the point.
  • the way of the inner difference viewpoint is: the L image is regarded as the leftmost L viewpoint, and a plurality of L viewpoint images are estimated according to the L image and the depth map; the R image is regarded as the rightmost R viewpoint, and according to the R image and the depth map Estimate multiple R viewpoint images.
  • the method of extending the viewpoint is to use the L image and the R image as the most intermediate pair of viewpoints; and estimating a plurality of L viewpoint images according to the L image and the depth map and the left and right directions, and estimating the left and right directions according to the R image and the depth map. Multiple R viewpoint images are output.
  • the viewpoint synthesis is performed, and according to the physical pixel arrangement of the ultra-high-definition naked-eye 3D screen, a plurality of viewpoint images are interleaved to form a pair of images; the image processed by the viewpoint synthesis includes a plurality of viewpoint information, and the number of pixels of the composite image is H. *V, where the value of H/M is an integer and the value of V/N is an integer.
  • the woven output is performed, and the composite image including the plurality of viewpoint information is format-woven, woven into a VBO (V-by-One) signal format, and output at a frequency of f2.
  • VBO V-by-One
  • the output VBO signal format is recorded as a second 3D video signal, and in step S400, the second 3D video signal is further processed.
  • step S400 the composite image is subjected to frame multiplication and image calibration, and transmitted to the naked eye 3D display to realize naked eye 3D display.
  • the step S400 specifically includes: S410, de-weaving the image signal of the VBO signal format to obtain a composite image including multiple viewpoints; performing frame interpolation on the composite image and Image calibration, the calibrated composite image is woven into a corresponding VBO signal format image signal and sent to a naked-eye 3D display to achieve naked-eye 3D display.
  • the VBO signal format image signal (that is, the second 3D video signal) is de-knitted to obtain a composite image including multiple viewpoints;
  • the image is subjected to frame interpolation and multiplication processing, and one frame of image is repeated to become the same T frame image, thereby obtaining a composite image of frequency Tf2; image calibration of the composite image, including calibration of Gamma, signal amplitude, etc.; and further
  • the calibrated composite image is woven into a VBO signal format and transmitted to the ultra-high definition naked-eye 3D screen at a frequency of Tf2 to realize high-definition naked-eye 3D display.
  • T is greater than 1, and T is a positive integer.
  • the step S410 includes:
  • S411 De-weaving the image signal of the VBO signal format, and converting the image signal of the VBO signal format into a composite image of the RGB signal format;
  • the calibrated composite image is woven into a corresponding VBO signal format image signal, and transmitted to the naked-eye 3D display at a frequency of Tf2 to realize naked-eye 3D display.
  • the above steps are a refinement of steps S400 and S410 of the present invention.
  • FIG. 5 is a schematic flowchart of a method of step S410 in a second embodiment of a naked eye 3D display method according to the present invention.
  • the de-knit is first performed: receiving the second 3D video signal, and converting the signal of the VBO transmission format into a composite image of the RGB signal format.
  • the frame multiplication is performed: the composite image is subjected to frame interpolation and frequency multiplication processing, and one frame image is repeated to become the same T frame image, thereby obtaining a composite image of frequency Tf2 (T is greater than 1).
  • the braided output is performed: the calibrated composite image is woven into a VBO signal format, and transmitted to the ultra-high-definition naked-eye 3D screen at a frequency of Tf2 to realize high-definition naked-eye 3D display.
  • the VBO signal format image signal outputted in step S400 can be recorded as a third 3D video signal.
  • the naked-eye 3D display receives the composite image including the plurality of viewpoint information in the VBO signal format outputted in step S400, thereby realizing high-definition naked-eye 3D display.
  • the present invention can implement the naked-eye 3D display by using the method flow shown in FIG. 1.
  • the method flow shown in FIG. 2 is a further explanation of the method flow shown in FIG. 1, that is, the second embodiment is A further refinement of the first embodiment will be described.
  • the image signal limitation of a specific format such as an LVDS signal format image signal and a VBO signal format image signal
  • this is for facilitating the transmission and reception of image data
  • those skilled in the art can also transmit images by using other formats of image signals after studying the present invention.
  • the data thus achieves the object of the invention, and therefore such modifications are within the scope of the appended claims.
  • the invention provides a naked-eye 3D display method, which performs image lifting and frame interpolation processing on the left and right eye images output by the 3D signal source device, and then weaves the left and right images and their identification signals into LVDS signals; acquires depth signals, and uses interpolation views.
  • the viewpoint generating a plurality of left-eye viewpoints by using the left-eye image and the depth information, generating a plurality of right-eye viewpoints by using the right-eye image and the depth information, and interlacing the plurality of left and right viewpoints into a VBO signal;
  • the VBO signal is processed by the frame multiplier, the VBO signal is also output to the super HD naked-eye 3D screen, driving ultra-high-definition naked-eye 3D screen to achieve high-definition naked-eye 3D display, which effectively realizes the conversion of naked-eye 3D viewpoint, consumes less resources, high definition of 3D display, smooth and smooth playback, low cost and wide application.
  • the present invention further provides a naked-eye 3D display system based on the above-described naked-eye 3D display method. As shown in FIG. 6, the system includes:
  • the decoding module 10 is configured to decode the 3D signal to obtain the original 3D video signal; specifically, as described in step S100;
  • the first 3D video processing module 20 is configured to perform signal separation, image enhancement, and classification into the original 3D video signal to obtain a corresponding L image and an R image; specifically, as described in step S200;
  • the second 3D video processing module 30 is configured to acquire depth information of the object object in the L image and the R image and estimate depth information of each pixel to form a corresponding depth map; and estimate the image according to the L image and the corresponding depth map.
  • a plurality of R view images are estimated according to the R image and the corresponding depth map; and the plurality of L view images and the plurality of R view images are interleaved to form a composite image including a plurality of views; specifically, step S300 Said
  • the third 3D video processing module 40 is configured to perform frame interpolation and image calibration on the composite image, and send the image to the naked eye 3D display to implement the naked eye 3D display; specifically, as described in step S400.
  • the first 3D video processing module 20 includes:
  • a first 3D video processing unit configured to perform signal separation on the original 3D video signal to obtain a corresponding L image signal and an R image signal; and to the L image signal and the R image
  • the image is image-enhanced and classified into a frame to obtain a corresponding L image sequence and an R image sequence.
  • the L image sequence and the R image sequence are format-woven to obtain a corresponding LVDS signal format image signal, as described in step S210.
  • the first 3D video processing unit includes:
  • a signal separation unit configured to perform signal separation on the original 3D video signal, and convert the original 3D image signal into a left image signal, a right image signal, and an L/R synchronization signal corresponding to the left image signal and the right image;
  • the color conversion and image lifting unit is configured to convert the color space image of the L image and the R image into a YUV color space, and perform image enhancement and brightness conversion on the L image and the R image of the YUV color space by using luminance partition scaling and hierarchical interpolation.
  • the classification inserting unit is configured to classify the L image and the R image with the pixel M*N under the control of the L/R synchronization signal, sequentially cache the L image together to form an L image set, and cache the R image together.
  • R image set respectively inserting L image set and R image set into frame, forming L image set with frame frequency f1 and R image set with frame frequency f1; sequentially outputting images of L image set and R image set, and Output corresponding to the R identification signal and the L identification signal; wherein, f1 is a positive integer;
  • An RGB image conversion unit for performing RGB image conversion on the L image and the R image, and converting the L image and the R image of the YUV color space into the L image and the R image of the RGB color space;
  • a first woven output unit for regularly arranging pixels in the L image and the R image of the RGB color space, and arranging the R, G, and B components in the pixel into LVDS according to rules
  • the signal format, and the L flag is inserted into the four pixels of the earliest transmission of the L picture, and the R flag is inserted into the four pixels of the earliest transmission of the R picture, and alternately output at a frame rate of 2f1.
  • the color conversion and image upgrading unit includes:
  • a color conversion unit for converting a color space image of the L image and the R image into a YUV color space, and converting pixels of the L image and the R image into pixels of YUV 4:2:2;
  • the brightness classification unit is configured to analyze and classify the brightness Y in the image signal of the YUV color space, and divide the brightness value into a low brightness area, a medium brightness area and a high brightness area according to the brightness value; the brightness of the low brightness area is less than the first critical a value, the brightness of the medium brightness region is between a first threshold value and a second threshold value, the brightness of the high brightness area is greater than a second threshold value; the first threshold value is less than a second threshold value;
  • the partitioning and scaling unit is configured to use brightness partitioning to enhance the contrast, to perform brightness compression on the image in the low-light area, to enhance the brightness of the image in the medium-light area, and to maintain the brightness of the high-brightness area unchanged;
  • the hierarchical interpolation unit is used to increase the resolution by using hierarchical interpolation, and different interpolation algorithms are used for different brightness regions to convert the image resolution into a fixed resolution, and the pixels of the converted L image and the R image are M*N arrays. .
  • the second 3D video processing module 30 includes:
  • a second 3D video processing unit configured to perform de-weaving and image separation on the LVDS signal format image signal, and separate the L image sequence and the R image sequence; and obtain depth information and estimation of the object object in the L image sequence and the R image sequence Depth information of each pixel to form a continuous smooth depth map; estimate from the L image and the corresponding depth map Extracting a plurality of L view images, estimating a plurality of R view images according to the R image and the corresponding depth map; and performing interleaving processing on the plurality of L view images and the plurality of R view images to form a composite image including the plurality of views;
  • the composite image is format-woven to obtain a corresponding VBO signal format image signal, as described in step S310.
  • the second 3D video processing unit includes:
  • a de-weaving and image separating unit for converting an LVDS signal format image signal into an RGB image signal, separating the L image and the R image;
  • a depth map generating unit for acquiring similar object objects in the L image and the R image, and calculating a parallax of the similar object object in the L image and the R image according to the horizontal displacement of the similar object object in the L image and the R image; And processing the color, texture and light and shadow information of the R image object, estimating the depth information of each pixel, and forming a continuous smooth depth map;
  • a multi-viewpoint generating unit configured to estimate a plurality of L view images according to the L image and the depth map by using an interpolation view or an epitaxial view, and estimate a plurality of R view images according to the R image and the depth map;
  • a view synthesis unit configured to perform a process of interleaving a plurality of L view images and a plurality of R view images according to a physical pixel arrangement of the naked eye 3D display, and synthesizing a composite image including the plurality of view information
  • a second woven output unit configured to format the composite image including the plurality of viewpoint information, woven into a corresponding VBO signal format image signal, and output at a frequency of f2, wherein the f2 is a positive integer.
  • the third 3D video processing module 40 includes:
  • a de-weaving unit for de-weaving the image signal of the VBO signal format, and converting the image signal of the VBO signal format into a composite image of the RGB signal format;
  • the frame interpolation frequency multiplying unit is configured to perform frame interpolation and frequency multiplication processing on the composite image, and repeat one frame image to become the same T frame image, thereby obtaining a composite image of frequency Tf2, wherein T is greater than 1;
  • An image calibration unit for performing image calibration on the composite image according to the electrical characteristics and display effect requirements of the naked eye 3D display
  • the woven output and display unit is configured to woven the calibrated composite image into a corresponding VBO signal format image signal, and transmit the signal to the naked-eye 3D display at a frequency of Tf2 to realize naked-eye 3D display.
  • the 3D signal output by the 3D signal source device is L/R format 720P@24Hz signal
  • the first 3D video signal is 1920 ⁇ 1080@120Hz signal
  • the second 3D video signal is 3840 ⁇ 2160@30Hz signal
  • the ultra high definition screen is 3840. ⁇ 2160@60Hz.
  • unit time (1s) is For 24 frames, the pixel resolution of the video frame is 1366 ⁇ 768, each frame is a flat image of the L image and the R image, and the pixel resolution of the L image and the R image is 683 ⁇ 768.
  • the first 3D video processing module respectively obtains independent L images and R images from the original 3D video signals, and the pixel resolutions of the L images and the R images are 683 ⁇ 768, respectively performing color conversion and brightness partition scaling on the L images and the R images.
  • hierarchical interpolation, conversion imaging An L image with a resolution of 1920 ⁇ 1080 and an R image with a pixel resolution of 1920 ⁇ 1080; and an L/R synchronization signal generated according to the original 3D video signal as shown in b) of FIG. 7: the synchronization signal is a high point It is an L image and an R image when it is low.
  • the L/R synchronization signal when the L/R synchronization signal is high level, it is an L image, and the L image is buffered together to form an L image set.
  • the L/R synchronization signal When the L/R synchronization signal is low level, it is an R image, and the R image is buffered.
  • the R image set is formed together, 24 L images in the L image set per unit time (1 s) and 24 R images in the R image set; the L image in the L image set is inserted as a reference, so that the L image per unit time is made.
  • the concentrated L image becomes 60 pairs, which also causes the R image in the R image set per unit time to become 60 pairs; and the generation frequency is 60 Hz synchronization signal, as shown in c in FIG.
  • the image of the L image set and the R image set are sequentially output, that is, the synchronous signal is high level to output L image, and the synchronous signal is low level to output R image, and the first 3D video signal with frame frequency of 120 Hz is formed as shown in FIG. 7 .
  • the frame frequency of the L image and the R image is 60 Hz.
  • the first 3D video processing module performs RGB image conversion on the 120Hz sequence frame, and then weaves into an LVDS (low voltage differential signal) transmission format, and inserts the L mark in the earliest transmitted four pixels of the L image, and the earliest in the R image The R flag is inserted into the four pixels transmitted, and finally transmitted to the second 3D video processing module at a frame rate of 120 Hz.
  • LVDS low voltage differential signal
  • the group LVDS signal line transmits pixels P1, P5, ..., Ptot-3,
  • the second group of LVDS signal lines transmits pixels P2, P6, ..., Ptot-2,
  • the third group of LVDS signal lines transmits pixels P3, P7, ..., Ptot-1,
  • the fourth group of LVDS signal lines transmits pixels P4, P8, ..., Ptot;
  • the pixel structure of the transmission is as shown in (b) of 8: the first four pixels P1, P2, P3, and P4 are transmitted including RGB image data, Hsyn and Vsyn synchronization data, L/R identification, And other data, other pixels have no L/R logo.
  • the second 3D video processing module receives the first 3D video signal, performs processing, separates the L image and the R image by using the L/R identifier, acquires the disparity between the L image and the R image, and further estimates the depth information between the pixels, A continuous smooth depth map is formed. And using a method of interpolating a viewpoint or an epitaxial viewpoint, estimating a plurality of L viewpoint images from the L image and the depth map, and estimating a plurality of R viewpoint images from the R image and the depth map to form a plurality of viewpoint images.
  • a plurality of viewpoint images are interleaved to form a composite image including a plurality of viewpoint information, and the number of pixels of the composite image is 3840 ⁇ 2160; and further synthesis of the multi-viewpoint information is included.
  • the image is format woven, woven into a VBO (V-by-One) signal format, and transmitted to the third 3D video processing module at a frequency of 30 Hz (shown as d in Figure 7).
  • the multi-view image is as shown in FIG. 9: the L image is the leftmost L viewpoint, and the L+1 + depth map inserts the L+1 viewpoint and the L+2 viewpoint; the same R image As the rightmost R viewpoint, and the R image + depth map inserts the R-1 viewpoint and the R-2 viewpoint; the six viewpoints are interlaced into a composite image of 3840 ⁇ 2160 according to the arrangement of the physical pixels of the ultra HD naked eye 3D screen, And transmitting to the third 3D video processing module at a frequency of 30 Hz; the third 3D video processing module performs processing such as inserting frame multiplication, The conversion frequency is 60Hz to drive the ultra-high-definition naked-eye 3D screen; the ultra-high-definition naked-eye 3D screen projects light, forming the L, R-2, L+1, R-1, L+2, R arranged in sequence as shown in FIG. Viewpoint image for high definition naked eye 3D display.
  • the present invention provides a naked-eye 3D system.
  • the first 3D video processing module performs image enhancement and frame interpolation on the left and right eye images output by the 3D signal source device, the left and right images and their identification signals are woven into LVDS signals for transmission to the first
  • the second 3D video processing module acquires the depth signal, and uses the left-eye image and the depth information to generate a plurality of left-eye viewpoints, and generates the right-eye image and the depth information by using the interpolation viewpoint or the epitaxial viewpoint.
  • a plurality of right-eye viewpoints are arranged by interlacing a plurality of left and right viewpoints and then woven into a VBO signal for transmission to a third 3D video processing module; and the third 3D video processing module performs processing such as frame multiplication and the like, and outputs the VBO signal to the ultra-HD naked eye.
  • 3D screen driving ultra-high definition naked-eye 3D screen to achieve high-definition naked-eye 3D display.
  • the invention can efficiently realize the conversion of the naked eye 3D viewpoint, consumes less resources, has high definition of 3D display, smooth and smooth playback, low cost and wide application.
  • the invention adopts brightness partition scaling and hierarchical interpolation to perform image lifting, and performs high-efficiency image processing while saving naked-eye 3D television resources, thereby improving image clarity of naked-eye 3D display.
  • the L/R sync mark is woven with the L/R image signal into an LVDS signal for transmission, saving the connection line and the CPU I/O port, saving resources.
  • the invention forms a depth map by acquiring the disparity of the L/R image, and then estimating the depth information of each pixel; forming a plurality of L viewpoints through the L image + depth map by using the interpolation view or the epitaxial viewpoint, and passing the R image + the depth map A plurality of R viewpoints are formed.
  • the invention adopts an ultra-high definition naked-eye 3D screen to realize high-definition naked-eye 3D display.
  • the present invention provides a naked-eye 3D display method and system, which obtains an original 3D video signal by decoding a 3D signal; performs signal separation, image enhancement, and classified interpolation on the original 3D video signal to obtain a corresponding L image and R image; acquiring depth information of the object object in the L image and the R image and estimating depth information of each pixel to form a corresponding depth map; estimating a plurality of L viewpoints according to the L image and the corresponding depth map An image, estimating a plurality of R view images according to the R image and the corresponding depth map; performing interleaving processing on the plurality of L view images and the plurality of R view images to form a composite image including the plurality of views; and inserting the composite image Frame multiplier and image calibration, and sent to the naked eye 3D display corresponding display; efficient realization of the naked eye 3D viewpoint conversion, less resources, 3D display clarity, smooth and smooth playback, low cost, wide application, bringing a big Convenience.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
  • Controls And Circuits For Display Device (AREA)

Abstract

一种裸眼3D显示方法及系统，通过对3D信号进行解码得到原始3D视频信号（S100）；对所述原始3D视频信号进行信号分离、图像提升及分类插帧，得到对应的L图像和R图像（S200）；获取L图像和R图像中物体对象的深度信息并估算出每个像素点的深度信息，形成对应的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像（S300）；对所述合成图像进行插帧倍频及图像校准，并发送给裸眼3D显示器对应显示（S400）；实现了裸眼3D视点的高效转换、占用资源少，3D显示清晰度高、播放平稳流畅，成本低、应用广泛。

Description

一种裸眼3D显示方法及系统 技术领域
本发明涉及裸眼3D显示领域，尤其涉及的是一种裸眼3D显示方法及系统。
背景技术
一般来说，3D(three-dimensional，三维)图像通常由两路图像组成，对应于某一特定的观看角度在人眼中便可实现3D效果，形成立体感，让人眼看上去就像真的一样。根据这两路视频所包含的视差信息，可以生成新的观看角度所对应的图像，这是裸眼3D电视的关键所在。
裸眼3D显示技术是不通过任何工具就能让左右两只眼睛从显示屏幕上看到两幅具有视差的、有所区别的画面，将它们反射到大脑，人就会产生立体感。它也利用了人眼的视差原理，通过给观看者左右两眼分别送去不同的画面，从而达到立体的视觉效果。由于裸眼3D电视的观察者可以不佩戴眼镜来进行3D显示体验，符合3D显示的市场需求，具有较大的市场和商机。目前3D信号源设备输出的3D信号一般都是左右(L/R)2视点，而裸眼3D电视需要较多视点以便在大范围进行3D体验，因此需要将2个视点转换成多个视点，而视点转换需要占用额外的电视资源，以及视点转换会降低观看的清晰度 等。
因此，现有技术还有待于改进和发展。
发明内容
本发明要解决的技术问题在于，提供一种裸眼3D显示方法及系统，旨在解决现有3D视点转换需要占用额外的电视资源，降低清晰度的问题。
本发明解决技术问题所采用的技术方案如下：
一种裸眼3D显示方法，其中，包括以下步骤：
S1、对3D信号进行解码得到原始3D视频信号；
S2、对所述原始3D视频信号进行信号分离、图像提升及分类插帧，得到对应的L图像和R图像；
S3、获取L图像和R图像中物体对象的深度信息并估算出每个像素点的深度信息，形成对应的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；
S4、对所述合成图像进行插帧倍频及图像校准，并发送给裸眼3D显示器，以实现裸眼3D显示。
所述的裸眼3D显示方法，其中，所述步骤S2具体包括：
S21、对所述原始3D视频信号进行信号分离，得到对应的L图像信号和R图像信号；对L图像信号和R图像信号进行图像提升及 分类插帧，得到对应的L图像序列和R图像序列；对L图像序列和R图像序列进行格式编织得到对应的LVDS信号格式图像信号。
所述的裸眼3D显示方法，其中，所述步骤S21具体包括：
S211、对原始3D视频信号进行信号分离，将原始3D图像信号转换成左图像信号、右图像信号及与左图像信号和右图像对应的L/R同步信号；
S212、将L图像和R图像的色彩空间图像转换为YUV色彩空间，对YUV色彩空间的L图像和R图像采用亮度分区缩放和分级插值来进行图像提升，转换成像素为M*N阵列的L和R图像；其中，M和N为正整数；
S213、在L/R同步信号控制下，对像素为M*N的L图像和R图像进行分类，依次将L图像缓存在一起形成L图像集，将R图像缓存在一起形成R图像集；分别对L图像集和R图像集进行插帧，形成帧频为f1的L图像集和帧频为f1的R图像集；将L图像集和R图像集中的图像依次输出、以及输出对应R标识信号和L标识信号；其中，f1为正整数；
S214、对L图像和R图像进行RGB图像转换，将YUV色彩空间的L图像和R图像对应转换成RGB色彩空间的L图像和R图像；
S215、对RGB色彩空间的L图像和R图像中的像素按规律排列，对像素中的R、G、B分量按规则编织成LVDS信号格式，并在L图像的最早传输的四个像素中插入L标识，在R图像的最早传输的四个像素中插入R标识，以帧频为2f1的速度交替输出。
所述的裸眼3D显示方法，其中，所述步骤S212具体包括：
S2121、将L图像和R图像的色彩空间图像转换成YUV色彩空间，将L图像和R图像的像素转化成YUV4:2:2的像素；
S2122、对YUV色彩空间的图像信号中亮度Y进行分析归类，按照亮度值高低分为低亮度区域、中亮度区域及高亮度区域；所述低亮度区域的亮度小于第一临界值，所述中亮度区域的亮度在第一临界值和第二临界值之间，所述高亮度区域的亮度大于第二临界值；所述第一临界值小于第二临界值；
S2123、采用亮度分区缩放来提升对比度，对低亮区域的图像进行亮度压缩，对中亮区域的图像进行亮度增强，高亮度区域的亮度保持不变；
S2124、采用分级插值来提升分辨率，对不同的亮度区域来采用不同的插值算法，将图像分辨率转换为固定分辨率，转换后L图像和R图像的像素为M*N阵列。
所述的裸眼3D显示方法，其中，所述步骤S3具体包括：
S31、对所述LVDS信号格式图像信号进行解编织及图像分离，分离出L图像序列和R图像序列；获取L图像序列和R图像序列中物体对象的深度信息及估算出每个像素点的深度信息，形成连续平滑的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；对所述合成图像进行格式编织得到对应的VBO信号格式图像信号。
所述的裸眼3D显示方法，其中，所述步骤S31具体包括：
S311、将LVDS信号格式图像信号转换成RGB图像信号、分离出L图像和R图像；
S312、获取L图像和R图像中相似的物体对象，根据L图像和R图像中相似物体对象的水平位移，计算出L图像和R图像中相似物体对象的视差；根据L图像和R图像物体对象的色彩、纹理及光影信息进行处理，估算出每个像素点的深度信息，形成连续平滑的深度图；
S313、采用内插视点或外延视点的方式，根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；
S314、根据裸眼3D显示器的物理像素排列情况，对多个L视点图像及多个R视点图像进行交织处理后合成包括多个视点信息的合成图像；
S315、对包含多个视点信息的合成图像进行格式编织，将其编织成对应的VBO信号格式图像信号，以频率为f2的速度输出，所述f2为正整数。
所述的裸眼3D显示方法，其中，所述步骤S4包括：
S41、对所述VBO信号格式图像信号进行解编织，将VBO信号格式的图像信号转换成RGB信号格式的合成图像；
S42、对合成图像进行插帧和倍频处理，将一帧图像进行重复后变为相同的T帧图像，从而得到频率为Tf2的合成图像，其中，T大 于1；
S43、根据裸眼3D显示器的电气特性及显示效果要求，对合成图像进行图像校准；
S44、将校准后的合成图像编织为对应的VBO信号格式图像信号，以频率为Tf2的速度传输给裸眼3D显示器，以实现裸眼3D显示。
一种裸眼3D显示系统，其中，包括：
解码模块，用于对3D信号进行解码得到原始3D视频信号；
第一3D视频处理模块，用于对所述原始3D视频信号进行信号分离、图像提升及分类插帧，得到对应的L图像和R图像；
第二3D视频处理模块，用于获取L图像和R图像中物体对象的深度信息并估算出每个像素点的深度信息，形成对应的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；
第三3D视频处理模块，用于对所述合成图像进行插帧倍频及图像校准，并发送给裸眼3D显示器，以实现裸眼3D显示。
所述的裸眼3D显示系统，其中，所述第一3D视频处理模块包括：
第一3D视频处理单元，用于对所述原始3D视频信号进行信号分离，得到对应的L图像信号和R图像信号；对L图像信号和R图像信号进行图像提升及分类插帧，得到对应的L图像序列和R图像 序列；对L图像序列和R图像序列进行格式编织得到对应的LVDS信号格式图像信号。
所述的裸眼3D显示系统，其中，所述第一3D视频处理单元包括：
信号分离单元，用于对原始3D视频信号进行信号分离，将原始3D图像信号转换成左图像信号、右图像信号及与左图像信号和右图像对应的L/R同步信号；
色彩转换及图像提升单元，用于将L图像和R图像的色彩空间图像转换成YUV色彩空间，对YUV色彩空间的L图像和R图像采用亮度分区缩放和分级插值来进行图像提升，转换成像素为M*N阵列的L和R图像；其中，M和N为正整数；
分类插帧单元，用于在L/R同步信号控制下，对像素为M*N的L图像和R图像进行分类，依次将L图像缓存在一起形成L图像集，将R图像缓存在一起形成R图像集；分别对L图像集和R图像集进行插帧，形成帧频为f1的L图像集和帧频为f1的R图像集；将L图像集和R图像集中的图像依次输出、以及输出对应R标识信号和L标识信号；其中，f1为正整数；
RGB图像转换单元，用于对L图像和R图像进行RGB图像转换，将YUV色彩空间的L图像和R图像对应转换成RGB色彩空间的L图像和R图像；
第一编织输出单元，用于对RGB色彩空间的L图像和R图像中的像素按规律排列，对像素中的R、G、B分量按规则编织成LVDS 信号格式，并在L图像的最早传输的四个像素中插入L标识，在R图像的最早传输的四个像素中插入R标识，以帧频为2f1的速度交替输出。
所述的裸眼3D显示系统，其中，所述色彩转换及图像提升单元包括：
色彩转换单元，用于将L图像和R图像的色彩空间图像转换成YUV色彩空间，将L图像和R图像的像素转化成YUV4:2:2的像素；
亮度分类单元，用于对YUV色彩空间的图像信号中亮度Y进行分析归类，按照亮度值高低分为低亮度区域、中亮度区域及高亮度区域；所述低亮度区域的亮度小于第一临界值，所述中亮度区域的亮度在第一临界值和第二临界值之间，所述高亮度区域的亮度大于第二临界值；所述第一临界值小于第二临界值；
分区缩放单元，用于采用亮度分区缩放来提升对比度，对低亮区域的图像进行亮度压缩，对中亮区域的图像进行亮度增强，高亮度区域的亮度保持不变；
分级插值单元，用于采用分级插值来提升分辨率，对不同的亮度区域来采用不同的插值算法，将图像分辨率转换为固定分辨率，转换后L图像和R图像的像素为M*N阵列。
所述的裸眼3D显示系统，其中，所述第二3D视频处理模块包括：
第二3D视频处理单元，用于对所述LVDS信号格式图像信号进行解编织及图像分离，分离出L图像序列和R图像序列；获取L图 像序列和R图像序列中物体对象的深度信息及估算出每个像素点的深度信息，形成连续平滑的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；对所述合成图像进行格式编织得到对应的VBO信号格式图像信号。
所述的裸眼3D显示系统，其中，所述第二3D视频处理单元包括：
解编织及图像分离单元，用于将LVDS信号格式图像信号转换成RGB图像信号、分离出L图像和R图像；
深度图生成单元，用于获取L图像和R图像中相似的物体对象，根据L图像和R图像中相似物体对象的水平位移，计算出L图像和R图像中相似物体对象的视差；根据L图像和R图像物体对象的色彩、纹理及光影信息进行处理，估算出每个像素点的深度信息，形成连续平滑的深度图；
多视点生成单元，用于采用内插视点或外延视点的方式，根据L图像和深度图估算出多个L视点图像，根据R图像和深度图估算出多个R视点图像；
视点合成单元，用于根据裸眼3D显示器的物理像素排列情况，对多个L视点图像及多个R视点图像进行交织处理后合成包括多个视点信息的合成图像；
第二编织输出单元，用于对包含多个视点信息的合成图像进行格 式编织，将其编织成对应的VBO信号格式图像信号，以频率为f2的速度输出，所述f2为正整数。
所述的裸眼3D显示系统，其中，所述第三3D视频处理模块包括：
解编织单元，用于对所述VBO信号格式图像信号进行解编织，将VBO信号格式的图像信号转换成RGB信号格式的合成图像；
插帧倍频单元，用于对合成图像进行插帧和倍频处理，将一帧图像进行重复后变为相同的T帧图像，从而得到频率为Tf2的合成图像，其中，T大于1；
图像校准单元，用于根据裸眼3D显示器的电气特性及显示效果要求，对合成图像进行图像校准；
编织输出及显示单元，用于将校准后的合成图像编织为对应的VBO信号格式图像信号，以频率为Tf2的速度传输给裸眼3D显示器，以实现裸眼3D显示。
本发明所提供的一种裸眼3D显示方法及系统，有效地解决了现有3D视点转换需要占用额外的电视资源，降低清晰度的问题，通过对3D信号进行解码得到原始3D视频信号；对所述原始3D视频信号进行信号分离、图像提升及分类插帧，得到对应的L图像和R图像；获取L图像和R图像中物体对象的深度信息并估算出每个像素点的深度信息，形成对应的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包 含多个视点的合成图像；对所述合成图像进行插帧倍频及图像校准，并发送给裸眼3D显示器，以实现裸眼3D显示；高效实现了裸眼3D视点的转换、占用资源少，3D显示清晰度高、播放平稳流畅，成本低、应用广泛，带来了大大的方便。
附图说明
图1为本发明提供的裸眼3D显示方法第一实施例的流程图。
图2为本发明提供的裸眼3D显示方法第二实施例的方法流程图。
图3为本发明提供的裸眼3D显示方法第二实施例中步骤S210的方法流程示意图。
图4为本发明提供的裸眼3D显示方法第二实施例中步骤S310的方法流程示意图。
图5为本发明提供的裸眼3D显示方法第二实施例中步骤S410的方法流程示意图。
图6为本发明提供的裸眼3D显示系统较佳实施例的结构框图。
图7为本发明提供的裸眼3D显示系统应用实施例中3D视频信号示意图。
图8为本发明提供的裸眼3D显示系统应用实施例中LVDS信号格式示意图。
图9为本发明提供的裸眼3D显示系统应用实施例中裸眼3D显示示意图。
具体实施方式
本发明提供一种裸眼3D显示方法及系统，为使本发明的目的、技术方案及优点更加清楚、明确，以下参照附图并举实施例对本发明进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
请参阅图1，图1为本发明提供的裸眼3D显示方法第一实施例的流程图，如图所示，所述方法包括以下步骤：
步骤S100、对3D信号进行解码得到原始3D视频信号；
步骤S200、对所述原始3D视频信号进行信号分离、图像提升及分类插帧，得到对应的L图像和R图像；
步骤S300、获取L图像和R图像中物体对象的深度信息并估算出每个像素点的深度信息，形成对应的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；
步骤S400、对所述合成图像进行插帧倍频及图像校准，并发送给裸眼3D显示器，以实现裸眼3D显示。
下面结合具体的实施例对上述步骤进行详细的描述。
在步骤S100中，对3D信号进行解码得到原始3D视频信号。具体来说裸眼3D可应用于如电视、手机、平板电脑等具有显示屏的终端中。现有应用最广泛的还是裸眼3D电视，以裸眼3D电视为例对本发明说明如下。3D信号为3D电视信号，裸眼3D电视的接口电路 接收外部3D信号源设备输出的3D电视信号，对其进行幅度限制、格式识别等处理，将处理后的3D电视信号输出给信号解码部分；解码部分对3D电视信号进行解码得到原始3D视频信号。
在步骤S200中，对所述原始3D视频信号进行信号分离、图像提升及分类插帧，得到对应的L图像和R图像。具体来说，请一并参阅图1和图2，图2为本发明提供的裸眼3D显示方法第二实施例的方法流程图，如图2所示，所述步骤S200包括：S210、对所述原始3D视频信号进行信号分离，得到对应的L图像信号和R图像信号；对L图像信号和R图像信号进行图像提升及分类插帧，得到对应的L图像序列和R图像序列；对L图像序列和R图像序列进行格式编织得到对应的LVDS信号格式图像信号。
具体来说，对原始3D视频信号进行信号分离，得到原始L(左眼)图像信号和R(右眼)图像信号、L/R同步信号；对L和R图像信号进行色彩转换、采用亮度分区缩放和分级插值来进行图像提升，转换成像素为M*N阵列的L和R图像；对像素为M*N的L和R图像进行分类和插图，将其转换成频率为f1的L图像序列和频率为f1的R图像序列；对图像序列中的L图像和R图像进行RGB图像转换、图像中的像素由R、G、B组成，以及进一步对L图像序列和R图像序列进行格式编织，将L图像及其L标识编织成LVDS(Low Voltage Differential Signaling，低压差分信号技术接口)信号格式、同样将R图像及其R标识编织成LVDS信号格式，以帧频为2f1的速度交替输出。其中，M、N及f1为正整数，具体取值可进行设置。其中，L图 像表示左图像，R图像表示右图像，关于L图像、R图像及L/R标识在3D显示技术领域中乃常用技术词语表达。
优选地，所述步骤S210具体包括：
S211、对原始3D视频信号进行信号分离，将原始3D图像信号转换成左图像信号、右图像信号及与左图像信号和右图像对应的L/R同步信号；
S212、将L图像和R图像的色彩空间图像转换成YUV色彩空间，对YUV色彩空间的L图像和R图像采用亮度分区缩放和分级插值来进行图像提升，转换成像素为M*N阵列的L和R图像；其中，M和N为正整数；
S213、在L/R同步信号控制下，对像素为M*N的L图像和R图像进行分类，依次将L图像缓存在一起形成L图像集，将R图像缓存在一起形成R图像集；分别对L图像集和R图像集进行插帧，形成帧频为f1的L图像集和帧频为f1的R图像集；将L图像集和R图像集中的图像依次输出、以及输出对应R标识信号和L标识信号；其中，f1为正整数；
S214、对L图像和R图像进行RGB图像转换，将YUV色彩空间的L图像和R图像对应转换成RGB色彩空间的L图像和R图像；
S215、对RGB色彩空间的L图像和R图像中的像素按规律排列，对像素中的R、G、B分量按规则编织成LVDS信号格式，并在L图像的最早传输的四个像素中插入L标识，在R图像的最早传输的四个像素中插入R标识，以帧频为2f1的速度交替输出。
请参阅图3，图3为本发明提供的裸眼3D显示方法第二实施例中步骤S210的方法流程示意图，在实际应用时，接收到原始3D视频信号后，对原始3D视频信号进行信号分离，将原始3D图像信号转换成左(L)图像信号和右(R)图像信号、以及与L图像信号和R图像对应的L/R同步信号。
然后进行色彩转换：将L图像和R图像的色彩空间图像转换成YUV(颜色编码方法，此乃现有技术名词)色彩空间，即将L图像和R图像的像素P(此处像素P代表图像中任意一像素点)转化成YUV4:2:2的像素，节省后续图像处理的带宽等资源及提高处理效率。对YUV色彩空间的L图像和R图像采用亮度分区缩放和分级插值来进行图像提升，转换成像素为M*N阵列的L和R图像；其中，M和N为正整数。
再在L/R同步信号控制下，对L图像和R图像进行分类，依次将L图像缓存在一起形成L图像集、将R图像缓存在一起形成R图像集；分别对L图像集和R图像集进行插帧，形成帧频为f1的L图像集和帧频为f1的R图像集。将L图像集和R图像集中的图像依次输出、以及输出对应R和L标识信号。其中，f1为频率值，为正整数。
再对L图像和R图像进行RGB图像转换，将YUV色彩空间的图像转换成RGB色彩空间的图像，RGB图像中的像素由R、G、B组成。
然后进行编织输出，对L图像和R图像中的像素按规律排列， 以及对像素中的R、G、B分量按规则编织成LVDS(低压差分信号)传输格式，并在L图像的最早传输的四个像素中插入L标识、在R图像的最早传输的四个像素中插入R标识，最后以帧频为2f1的速度交替输出。如图3所示，可将输出的LVDS图像信号记为第一3D视频信号。然后对第一3D视频信号进行如步骤S300所述的处理。
优选地，在实际应用时，所述步骤S212具体包括：
S2121、将L图像和R图像的色彩空间图像转换成YUV色彩空间，将L图像和R图像的像素转化成YUV4:2:2的像素；
S2122、对YUV色彩空间的图像信号中亮度Y进行分析归类，按照亮度值高低分为低亮度区域、中亮度区域及高亮度区域；所述低亮度区域的亮度小于第一临界值，所述中亮度区域的亮度在第一临界值和第二临界值之间，所述高亮度区域的亮度大于第二临界值；所述第一临界值小于第二临界值；
S2123、采用亮度分区缩放来提升对比度，对低亮区域的图像进行亮度压缩，对中亮区域的图像进行亮度增强，高亮度区域的亮度保持不变；
S2124、采用分级插值来提升分辨率，对不同的亮度区域来采用不同的插值算法，将图像分辨率转换为固定分辨率，转换后L图像和R图像的像素为M*N阵列。
具体来说，对YUV色彩空间的L图像和R图像进行图像提升，包括以下步骤：
首先对YUV色彩空间的图像信号中亮度Y进行分析归类，分为： 低亮度区域(亮度小于第一临界值Yd)、高亮度区域(亮度大于第二临界值Yh)、中亮度区域(亮度在第一临界值Yd和第二临界值Yh之间)。其中，第一临界值Yd小于第二临界值Yh，具体可根据实际需要进行设置。
然后，采用亮度分区缩放来提升对比度：对低亮区域的图像进行亮度压缩、即低亮区域的像素的亮度Pd(Pd为低亮区域的像素的亮度值)变为Pd*Cd，Cd是变量、且Cd不大于1；对中亮区域的图像进行亮度增强、即中亮区域的像素的亮度Pz(Pz为中亮区域的像素的亮度值)变为Pz*Cz，Cz是变量、且Cz不小于1并且Pz*Cz不大于高亮区域的临界值Yh。而高亮度区域的亮度保持不变。
由于低亮度区域亮度降低、中亮度区域亮度抬高，使得对比度提升，同时高亮度区域亮度不变、保证图像不失真。
然后，采用分级插值来提升分辨率，即对不同的亮度区域来采用不同的插值算法，将图像分辨率转换为固定分辨率，转换后L图像和R图像的像素为M*N阵列。具体为：
低亮度区域和高亮度区域、采用四点平均算法来进行插值，即：设待插值像素Px周围相邻的四个像素分别为Pa(Ya,Ua,Va)、Pb(Yb,Ub,Vb)、Pm(Ym,Um,Vm)、Pn(Yn,Un,Vn)，则插入的像素Px为(Yx,Ux,Vx)，其中Yx＝(Ya+Yb+Ym+Yn)/4、Ux＝(Ua+Ub+Um+Un)/4、Vx＝(Va+Vb+Vm+Vn)/4。其中，Y，U，V等分别表示四个像素在YUV色彩空间的分量值，此乃现有技术，本领域普通技术人员很容易理解，不作赘述。
而中亮度区域采用紧密关联算法来进行插值，即待插值像素Px周围相邻的已知的四个像素分别为Pa(Ya,Ua,Va)、Pb(Yb,Ub,Vb)、Pm(Ym,Um,Vm)、Pn(Yn,Un,Vn)，则插入的像素Px为(Yx,Ux,Vx)。
其中，Yx、Ux、Vx计算过程为：以Yx值为例，
1)先计算相邻四个像素的亮度平均值Ye、Ye＝(Ya+Yb+Ym+Yn)/4；
2)分别计算出Ya、Yb、Ym、Yn与Ye的方差Wa、Wb、Wm和Wn,以及方差和W＝Wa+Wb+Wm+Wn；其中，Wa、Wb、Wm和Wn表示Ya、Yb、Ym、Yn与Ye对应的方差值；
3)计算权重系数Ka＝1-Wa/W、Kb＝1-Wb/W、Km＝1-Wm/W、Kn＝1-Wn/W；
4)Yx＝Ka×Ya+Kb×Yb+Km×Ym+Kn×Yn。Ux、Vx的计算过程与Yx相同，不再赘述。
在步骤S300中，获取L图像和R图像中物体对象的深度信息并估算出每个像素点的深度信息，形成对应的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像。
具体来说，请参阅图2，所述步骤S300具体包括：S310、对所述LVDS信号格式图像信号进行解编织及图像分离，分离出L图像序列和R图像序列；获取L图像序列和R图像序列中物体对象的深度信息及估算出每个像素点的深度信息，形成连续平滑的深度图；根据 L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；对所述合成图像进行格式编织得到对应的VBO信号格式图像信号。
在实际应用时，对所述LVDS信号格式图像信号(也就是第一3D视频信号)进行解编织及L/R图像分离处理，分离出L图像和R图像；对L图像和R图像进行处理，获取图像中物体对象的深度信息以及估算出每个像素点的深度信息，形成连续平滑的深度图；采用内插视点或外延视点的方式，根据L图像和深度图、估算出多个L视点图像，以及根据R图像和深度图、估算出多个R视点图像，形成多个视点图像；根据超高清裸眼3D屏的物理像素排列情况，对多个视点图像进行交织处理后形成包含多个视点信息的合成图像，合成图像的像素数为H*V，其中H/M的值为整数、V/N的值为整数。以及进一步对包含多视点信息的合成图像进行格式编织，将其编织成VBO(V-by-One，此乃现有技术名词)信号格式、以频率为f2的速度输出。
优选地，所述步骤S310具体包括：
S311、将LVDS信号格式图像信号转换成RGB图像信号、分离出L图像和R图像；
S312、获取L图像和R图像中相似的物体对象，根据L图像和R图像中相似物体对象的水平位移，计算出L图像和R图像中相似物体对象的视差；根据L图像和R图像物体对象的色彩、纹理及光 影信息进行处理，估算出每个像素点的深度信息，形成连续平滑的深度图；
S313、采用内插视点或外延视点的方式，根据L图像和深度图估算出多个L视点图像，根据R图像和深度图估算出多个R视点图像；
S314、根据裸眼3D显示器的物理像素排列情况，对多个L视点图像及多个R视点图像进行交织处理后合成包括多个视点信息的合成图像；
S315、对包含多个视点信息的合成图像进行格式编织，将其编织成对应的VBO信号格式图像信号，以频率为f2的速度输出，所述f2为正整数。
请参阅图4，图4为本发明提供的裸眼3D显示方法第二实施例中步骤S310的方法流程示意图。在实际应用时，先进行解编织及L/R图像分离：接收LVDS信号格式图像信号，将其LVDS传输格式的信号转换成RGB图像信号、并在此过程中获取到L标识或R标识，通过L标识分离出L图像、和通过R标识分离出R图像。
然后进行深度图生成：获取L图像和R图像中相似的物体对象，根据L图像和R图像中相似物体对象的水平位移，计算出L图像和R图像中相似物体对象的深度(或称为视差)；然后根据L图像和R图像对象的色彩、纹理、光影等信息进行处理，估算出每个像素点的深度信息，进一步形成连续平滑的深度图。
然后进行多视点生成：多视点生成可采用内插视点的方式、或外 延视点的方式。内差视点的方式即：将L图像作为最左边的L视点，并根据L图像和深度图、估算出多个L视点图像；将R图像作为最右边的R视点，并根据R图像和深度图、估算出多个R视点图像。外延视点的方式即：将L图像和R图像作为最中间的一对视点；并根据L图像和深度图、相左右方向估算出多个L视点图像，根据R图像和深度图、向左右方向估算出多个R视点图像。
然后进行视点合成，根据超高清裸眼3D屏的物理像素排列情况，对多个视点图像进行交织处理后合成一副图像；通过视点合成处理的图像包括多个视点信息，合成图像的像素数为H*V，其中H/M的值为整数、V/N的值为整数。
然后进行编织输出，对包含多个视点信息的合成图像进行格式编织，将其编织成VBO(V-by-One)信号格式、以频率为f2的速度输出。如图4所示，将输出的VBO信号格式记为第二3D视频信号，步骤S400是对第二3D视频信号再进行处理。
在步骤S400中，对所述合成图像进行插帧倍频及图像校准，并发送给裸眼3D显示器，以实现裸眼3D显示。具体来说，请参阅图2，所述步骤S400具体包括：S410、对所述VBO信号格式图像信号进行解编织，得到包含多个视点的合成图像；对所述合成图像进行插帧倍频及图像校准，将校准后的合成图像编织为对应的VBO信号格式图像信号并发送给裸眼3D显示器，以实现裸眼3D显示。
在实际应用时，对所述VBO信号格式图像信号(也就是第二3D视频信号)进行解编织处理，得到包含多个视点的合成图像；对合成 图像进行插帧和倍频处理，将一帧图像进行重复后变为相同的T帧图像，从而得到频率为Tf2的合成图像；对合成图像进行图像校准，包括Gamma、信号幅度等校准；以及进一步将校准后的合成图像编织为VBO信号格式、以频率为Tf2的速度传输给超高清裸眼3D屏，实现高清晰度裸眼3D显示。其中，T大于1，且T为正整数。
优选地，所述步骤S410包括：
S411、对所述VBO信号格式图像信号进行解编织，将VBO信号格式的图像信号转换成RGB信号格式的合成图像；
S412、对合成图像进行插帧和倍频处理，将一帧图像进行重复后变为相同的T帧图像，从而得到频率为Tf2的合成图像，其中，T大于1；
S413、根据裸眼3D显示器的电气特性及显示效果要求，对合成图像进行图像校准；
S414、将校准后的合成图像编织为对应的VBO信号格式图像信号，以频率为Tf2的速度传输给裸眼3D显示器，以实现裸眼3D显示。上述步骤是对本发明步骤S400及S410的细化。
请参阅图5，图5为本发明提供的裸眼3D显示方法第二实施例中步骤S410的方法流程示意图。在实际应用时，先进行解编织：接收第二3D视频信号，将其VBO传输格式的信号转换成RGB信号格式的合成图像。然后进行插帧倍频：对合成图像进行插帧和倍频处理，将一帧图像进行重复后变为相同的T帧图像，从而得到频率为Tf2的合成图像(T大于1)。然后，进行图像校准：根据超高清裸眼3D 屏的电气特性、以及显示效果的要求，对合成图像进行图像校准，包括Gamma、信号幅度等校准。最后，进行编织输出：将校准后的合成图像编织为VBO信号格式、以频率为Tf2的速度传输给超高清裸眼3D屏，实现高清晰度裸眼3D显示。如图5所示，可将步骤S400中编织输出的VBO信号格式图像信号记为第三3D视频信号。裸眼3D显示器(譬如超高清裸眼3D屏)接收步骤S400输出的VBO信号格式的包含多个视点信息的合成图像，便可实现高清晰度裸眼3D显示。
需要说明的是，本发明采用图1所示的方法流程便可实现裸眼3D显示，图2所示的方法流程是对图1所示的方法流程的进一步阐述，也就是说第二实施例是对第一实施例的进一步细化说明。关于LVDS信号格式图像信号及VBO信号格式图像信号等具体格式的图像信号限定，这是为了方便图像数据的传输与接收，本领域普通技术人员在研究本发明之后还可采用其它格式图像信号传输图像数据从而实现本发明的目的，因此这些变换属于本发明所附权利要求的保护范围之内。
本发明提出一种裸眼3D显示方法，对3D信号源设备输出的左右眼图像进行图像提升和插帧等处理后，将左右图及其标识信号编织成LVDS信号；获取深度信号，采用内插视点或外延视点的方式，通过左眼图像及深度信息来生成多个左眼视点，通过右眼图像及深度信息来生成多个右眼视点，将多个左右视点交织排列后编织成VBO信号；对VBO信号进行插帧倍频等处理后同样以VBO信号输出给超 高清裸眼3D屏，驱动超高清裸眼3D屏来实现高清晰度裸眼3D显示，从而高效实现了裸眼3D视点的转换、占用资源少，3D显示清晰度高、播放平稳流畅，以及成本低、应用广泛。
基于上述裸眼3D显示方法，本发明还提供了一种裸眼3D显示系统，如图6所示，所述系统包括：
解码模块10，用于对3D信号进行解码得到原始3D视频信号；具体如步骤S100所述；
第一3D视频处理模块20，用于对所述原始3D视频信号进行信号分离、图像提升及分类插帧，得到对应的L图像和R图像；具体如步骤S200所述；
第二3D视频处理模块30，用于获取L图像和R图像中物体对象的深度信息并估算出每个像素点的深度信息，形成对应的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；具体如步骤S300所述；
第三3D视频处理模块40，用于对所述合成图像进行插帧倍频及图像校准，并发送给裸眼3D显示器，以实现裸眼3D显示；具体如步骤S400所述。
进一步地，所述第一3D视频处理模块20包括：
第一3D视频处理单元，用于对所述原始3D视频信号进行信号分离，得到对应的L图像信号和R图像信号；对L图像信号和R图 像信号进行图像提升及分类插帧，得到对应的L图像序列和R图像序列；对L图像序列和R图像序列进行格式编织得到对应的LVDS信号格式图像信号，具体如步骤S210所述。
进一步地，所述第一3D视频处理单元包括：
信号分离单元，用于对原始3D视频信号进行信号分离，将原始3D图像信号转换成左图像信号、右图像信号及与左图像信号和右图像对应的L/R同步信号；
色彩转换及图像提升单元，用于将L图像和R图像的色彩空间图像转换成YUV色彩空间，对YUV色彩空间的L图像和R图像采用亮度分区缩放和分级插值来进行图像提升，转换成像素为M*N阵列的L和R图像；其中，M和N为正整数；
分类插帧单元，用于在L/R同步信号控制下，对像素为M*N的L图像和R图像进行分类，依次将L图像缓存在一起形成L图像集，将R图像缓存在一起形成R图像集；分别对L图像集和R图像集进行插帧，形成帧频为f1的L图像集和帧频为f1的R图像集；将L图像集和R图像集中的图像依次输出、以及输出对应R标识信号和L标识信号；其中，f1为正整数；
RGB图像转换单元，用于对L图像和R图像进行RGB图像转换，将YUV色彩空间的L图像和R图像对应转换成RGB色彩空间的L图像和R图像；
第一编织输出单元，用于对RGB色彩空间的L图像和R图像中的像素按规律排列，对像素中的R、G、B分量按规则编织成LVDS 信号格式，并在L图像的最早传输的四个像素中插入L标识，在R图像的最早传输的四个像素中插入R标识，以帧频为2f1的速度交替输出。
进一步地，所述色彩转换及图像提升单元包括：
色彩转换单元，用于将L图像和R图像的色彩空间图像转换成YUV色彩空间，将L图像和R图像的像素转化成YUV4:2:2的像素；
亮度分类单元，用于对YUV色彩空间的图像信号中亮度Y进行分析归类，按照亮度值高低分为低亮度区域、中亮度区域及高亮度区域；所述低亮度区域的亮度小于第一临界值，所述中亮度区域的亮度在第一临界值和第二临界值之间，所述高亮度区域的亮度大于第二临界值；所述第一临界值小于第二临界值；
分区缩放单元，用于采用亮度分区缩放来提升对比度，对低亮区域的图像进行亮度压缩，对中亮区域的图像进行亮度增强，高亮度区域的亮度保持不变；
分级插值单元，用于采用分级插值来提升分辨率，对不同的亮度区域来采用不同的插值算法，将图像分辨率转换为固定分辨率，转换后L图像和R图像的像素为M*N阵列。
进一步地，所述第二3D视频处理模块30包括：
第二3D视频处理单元，用于对所述LVDS信号格式图像信号进行解编织及图像分离，分离出L图像序列和R图像序列；获取L图像序列和R图像序列中物体对象的深度信息及估算出每个像素点的深度信息，形成连续平滑的深度图；根据L图像和对应的深度图估算 出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；对所述合成图像进行格式编织得到对应的VBO信号格式图像信号，具体如步骤S310所述。
进一步地，所述第二3D视频处理单元包括：
解编织及图像分离单元，用于将LVDS信号格式图像信号转换成RGB图像信号、分离出L图像和R图像；
深度图生成单元，用于获取L图像和R图像中相似的物体对象，根据L图像和R图像中相似物体对象的水平位移，计算出L图像和R图像中相似物体对象的视差；根据L图像和R图像物体对象的色彩、纹理及光影信息进行处理，估算出每个像素点的深度信息，形成连续平滑的深度图；
多视点生成单元，用于采用内插视点或外延视点的方式，根据L图像和深度图估算出多个L视点图像，根据R图像和深度图估算出多个R视点图像；
视点合成单元，用于根据裸眼3D显示器的物理像素排列情况，对多个L视点图像及多个R视点图像进行交织处理后合成包括多个视点信息的合成图像；
第二编织输出单元，用于对包含多个视点信息的合成图像进行格式编织，将其编织成对应的VBO信号格式图像信号，以频率为f2的速度输出，所述f2为正整数。
进一步地，所述第三3D视频处理模块40包括：
解编织单元，用于对所述VBO信号格式图像信号进行解编织，将VBO信号格式的图像信号转换成RGB信号格式的合成图像；
插帧倍频单元，用于对合成图像进行插帧和倍频处理，将一帧图像进行重复后变为相同的T帧图像，从而得到频率为Tf2的合成图像，其中，T大于1；
图像校准单元，用于根据裸眼3D显示器的电气特性及显示效果要求，对合成图像进行图像校准；
编织输出及显示单元，用于将校准后的合成图像编织为对应的VBO信号格式图像信号，以频率为Tf2的速度传输给裸眼3D显示器，以实现裸眼3D显示。
以下以一应用实施例对本发明详细说明如下。
假设3D信号源设备输出的3D信号为L/R格式720P@24Hz信号、第一3D视频信号为1920×1080@120Hz信号、第二3D视频信号为3840×2160@30Hz信号、超高清屏幕为3840×2160@60Hz。
请参阅图7，3D信号源设备输出的L/R 720P@24Hz 3D视频信号，经过接口电路及解码模块处理后输出的原始3D视频信号如图7中a)所示：单位时间(1s)为24帧，视频帧的像素分辨率为1366×768，每一帧是L图像和R图像的拼合图像，L图像和R图像的像素分辨率为683×768。
第一3D视频处理模块从原始3D视频信号中分别获取独立的L图像和R图像，L图像和R图像的像素分辨率为683×768，分别对L图像和R图像进行色彩转换、亮度分区缩放和分级插值，转换成像 素分辨率为1920×1080的L图像和像素分辨率为1920×1080的R图像；以及根据原始3D视频信号来生成L/R同步信号如图7中b)所示：同步信号为高点平时为L图像、为低电平时为R图像。
进一步进行分类插帧：在L/R同步信号高电平时则为L图像、将L图像缓存在一起形成L图像集，在L/R同步信号低电平时则为R图像、将R图像缓存在一起形成R图像集，单位时间(1s)的L图像集中有24副L图像、R图像集中有24副R图像；以L图像集中的L图像为参照进行插帧、使得单位时间内的L图像集中的L图像变为60副，同样使得单位时间内的R图像集中的R图像变为60副；以及生成频率为60Hz同步信号，如图7中c)所示；在同步信号的控制下交替依次输出L图像集和R图像集中的图像，即同步信号为高电平则输出L图像、同步信号为低电平则输出R图像，形成了帧频为120Hz的第一3D视频信号如图7中c)所示，其中L图像和R图像的帧频为60Hz。
以及第一3D视频处理模块对120Hz的序列帧进行RGB图像转换后，编织成LVDS(低压差分信号)传输格式，并在L图像的最早传输的四个像素中插入L标识、在R图像的最早传输的四个像素中插入R标识，最后以帧频为120Hz的速度传输给第二3D视频处理模块。
请参阅图8，其中，第一3D视频处理模块采用四组LVDS信号线，编织的LVDS传输信号如图8中(a)所示：每一帧的像素数为tot＝1920×1080，第一组LVDS信号线传输像素P1、P5、...、Ptot-3， 第二组LVDS信号线传输像素P2、P6、...、Ptot-2，第三组LVDS信号线传输像素P3、P7、...、Ptot-1，第四组LVDS信号线传输像素P4、P8、...、Ptot；传输的像素结构如8中(b)所示：最早传输的四个像素P1、P2、P3、P4包括RGB图像数据、Hsyn和Vsyn同步数据、L/R标识、及其它数据，其它像素没有L/R标识。
第二3D视频处理模块接收第一3D视频信号后进行处理，通过L/R标识来分离出L图像和R图像，获取L图像和R图像间的视差、以及进一步估算各个像素间的深度信息、形成连续平滑的深度图。以及采用内插视点或外延视点的方式，根据L图像和深度图、估算出多个L视点图像，以及根据R图像和深度图、估算出多个R视点图像，形成多个视点图像。根据超高清裸眼3D屏的物理像素排列情况，对多个视点图像进行交织处理后形成包含多个视点信息的合成图像，合成图像的像素数为3840×2160；以及进一步对包含多视点信息的合成图像进行格式编织，将其编织成VBO(V-by-One)信号格式、以频率为30Hz的速度(如图7中d)所示)传输给第三3D视频处理模块。
以6个视点内插视点为例，多视点图像如图9所示：L图像作为最左边的L视点，并且L图像+深度图内插出L+1视点和L+2视点；同样R图像作为最右的R视点，并且R图像+深度图内插出R-1视点和R-2视点；6个视点按照超高清裸眼3D屏的物理像素的排列来交织成3840×2160的合成图像，并以频率为30Hz的速度传输给第三3D视频处理模块；第三3D视频处理模块进行插帧倍频等处理， 转换频率为60Hz的速度来驱动超高清裸眼3D屏；超高清裸眼3D屏投射出光线，形成图9所示依次排列的L、R-2、L+1、R-1、L+2、R视点图像，实现高清晰度裸眼3D显示。
本发明提出一种裸眼3D系统，第一3D视频处理模块对3D信号源设备输出的左右眼图像进行图像提升和插帧等处理后，将左右图及其标识信号编织成LVDS信号一起传输给第二3D视频处理模块；第二3D视频处理模块获取深度信号，采用内插视点或外延视点的方式，通过左眼图像及深度信息来生成多个左眼视点、通过右眼图像及深度信息来生成多个右眼视点，将多个左右视点交织排列后编织成VBO信号传输给第三3D视频处理模块；第三3D视频处理模块进行插帧倍频等处理后同样以VBO信号输出给超高清裸眼3D屏，驱动超高清裸眼3D屏来实现高清晰度裸眼3D显示。本发明能够高效实现裸眼3D视点的转换、占用资源少，3D显示清晰度高、播放平稳流畅，以及成本低、应用广泛。
本发明采用亮度分区缩放和分级插值来进行图像提升，在节省裸眼3D电视资源的情况下进行高效率图像处理，提高裸眼3D显示的图像清晰度。将L/R同步标识与L/R图像信号一起编织成LVDS信号进行传输，节省连接线以及CPU I/O口，节省资源。本发明通过获取L/R图像的视差，然后估算各个像素的深度信息形成深度图；采用内插视点或外延视点的方式，通过L图像+深度图形成多个L视点、通过R图像+深度图形成多个R视点。本发明采用超高清裸眼3D屏，实现高清晰度裸眼3D显示。
综上所述，本发明提供的一种裸眼3D显示方法及系统，通过对3D信号进行解码得到原始3D视频信号；对所述原始3D视频信号进行信号分离、图像提升及分类插帧，得到对应的L图像和R图像；获取L图像和R图像中物体对象的深度信息并估算出每个像素点的深度信息，形成对应的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；对所述合成图像进行插帧倍频及图像校准，并发送给裸眼3D显示器对应显示；高效实现了裸眼3D视点的转换、占用资源少，3D显示清晰度高、播放平稳流畅，成本低、应用广泛，带来了大大的方便。
应当理解的是，本发明的应用不限于上述的举例，对本领域普通技术人员来说，可以根据上述说明加以改进或变换，所有这些改进和变换都应属于本发明所附权利要求的保护范围。

Claims (14)

  1. 一种裸眼3D显示方法，其特征在于，包括以下步骤：
    S1、对3D信号进行解码得到原始3D视频信号；
    S2、对所述原始3D视频信号进行信号分离、图像提升及分类插帧，得到对应的L图像和R图像；
    S3、获取L图像和R图像中物体对象的深度信息并估算出每个像素点的深度信息，形成对应的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；
    S4、对所述合成图像进行插帧倍频及图像校准，并发送给裸眼3D显示器，以实现裸眼3D显示。
  2. 根据权利要求1所述的裸眼3D显示方法，其特征在于，所述步骤S2具体包括：
    S21、对所述原始3D视频信号进行信号分离，得到对应的L图像信号和R图像信号；对L图像信号和R图像信号进行图像提升及分类插帧，得到对应的L图像序列和R图像序列；对L图像序列和R图像序列进行格式编织得到对应的LVDS信号格式图像信号。
  3. 根据权利要求2所述的裸眼3D显示方法，其特征在于，所述步骤S21具体包括：
    S211、对原始3D视频信号进行信号分离，将原始3D图像信号转换成左图像信号、右图像信号及与左图像信号和右图像对应的L/R 同步信号；
    S212、将L图像和R图像的色彩空间图像转换为YUV色彩空间，对YUV色彩空间的L图像和R图像采用亮度分区缩放和分级插值来进行图像提升，转换成像素为M*N阵列的L和R图像；其中，M和N为正整数；
    S213、在L/R同步信号控制下，对像素为M*N的L图像和R图像进行分类，依次将L图像缓存在一起形成L图像集，将R图像缓存在一起形成R图像集；分别对L图像集和R图像集进行插帧，形成帧频为f1的L图像集和帧频为f1的R图像集；将L图像集和R图像集中的图像依次输出、以及输出对应R标识信号和L标识信号；其中，f1为正整数；
    S214、对L图像和R图像进行RGB图像转换，将YUV色彩空间的L图像和R图像对应转换成RGB色彩空间的L图像和R图像；
    S215、对RGB色彩空间的L图像和R图像中的像素按规律排列，对像素中的R、G、B分量按规则编织成LVDS信号格式，并在L图像的最早传输的四个像素中插入L标识，在R图像的最早传输的四个像素中插入R标识，以帧频为2f1的速度交替输出。
  4. 根据权利要求3所述的裸眼3D显示方法，其特征在于，所述步骤S212具体包括：
    S2121、将L图像和R图像的色彩空间图像转换成YUV色彩空间，将L图像和R图像的像素转化成YUV4:2:2的像素；
    S2122、对YUV色彩空间的图像信号中亮度Y进行分析归类， 按照亮度值高低分为低亮度区域、中亮度区域及高亮度区域；所述低亮度区域的亮度小于第一临界值，所述中亮度区域的亮度在第一临界值和第二临界值之间，所述高亮度区域的亮度大于第二临界值；所述第一临界值小于第二临界值；
    S2123、采用亮度分区缩放来提升对比度，对低亮区域的图像进行亮度压缩，对中亮区域的图像进行亮度增强，高亮度区域的亮度保持不变；
    S2124、采用分级插值来提升分辨率，对不同的亮度区域来采用不同的插值算法，将图像分辨率转换为固定分辨率，转换后L图像和R图像的像素为M*N阵列。
  5. 根据权利要求2所述的裸眼3D显示方法，其特征在于，所述步骤S3具体包括：
    S31、对所述LVDS信号格式图像信号进行解编织及图像分离，分离出L图像序列和R图像序列；获取L图像序列和R图像序列中物体对象的深度信息及估算出每个像素点的深度信息，形成连续平滑的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；对所述合成图像进行格式编织得到对应的VBO信号格式图像信号。
  6. 根据权利要求5所述的裸眼3D显示方法，其特征在于，所述步骤S31具体包括：
    S311、将LVDS信号格式图像信号转换成RGB图像信号、分离 出L图像和R图像；
    S312、获取L图像和R图像中相似的物体对象，根据L图像和R图像中相似物体对象的水平位移，计算出L图像和R图像中相似物体对象的视差；根据L图像和R图像物体对象的色彩、纹理及光影信息进行处理，估算出每个像素点的深度信息，形成连续平滑的深度图；
    S313、采用内插视点或外延视点的方式，根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；
    S314、根据裸眼3D显示器的物理像素排列情况，对多个L视点图像及多个R视点图像进行交织处理后合成包括多个视点信息的合成图像；
    S315、对包含多个视点信息的合成图像进行格式编织，将其编织成对应的VBO信号格式图像信号，以频率为f2的速度输出，所述f2为正整数。
  7. 根据权利要求6所述的裸眼3D显示方法，其特征在于，所述步骤S4包括：
    S41、对所述VBO信号格式图像信号进行解编织，将VBO信号格式的图像信号转换成RGB信号格式的合成图像；
    S42、对合成图像进行插帧和倍频处理，将一帧图像进行重复后变为相同的T帧图像，从而得到频率为Tf2的合成图像，其中，T大于1；
    S43、根据裸眼3D显示器的电气特性及显示效果要求，对合成图像进行图像校准；
    S44、将校准后的合成图像编织为对应的VBO信号格式图像信号，以频率为Tf2的速度传输给裸眼3D显示器，以实现裸眼3D显示。
  8. 一种裸眼3D显示系统，其特征在于，包括：
    解码模块，用于对3D信号进行解码得到原始3D视频信号；
    第一3D视频处理模块，用于对所述原始3D视频信号进行信号分离、图像提升及分类插帧，得到对应的L图像和R图像；
    第二3D视频处理模块，用于获取L图像和R图像中物体对象的深度信息并估算出每个像素点的深度信息，形成对应的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；
    第三3D视频处理模块，用于对所述合成图像进行插帧倍频及图像校准，并发送给裸眼3D显示器，以实现裸眼3D显示。
  9. 根据权利要求8所述的裸眼3D显示系统，其特征在于，所述第一3D视频处理模块包括：
    第一3D视频处理单元，用于对所述原始3D视频信号进行信号分离，得到对应的L图像信号和R图像信号；对L图像信号和R图像信号进行图像提升及分类插帧，得到对应的L图像序列和R图像序列；对L图像序列和R图像序列进行格式编织得到对应的LVDS 信号格式图像信号。
  10. 根据权利要求9所述的裸眼3D显示系统，其特征在于，所述第一3D视频处理单元包括：
    信号分离单元，用于对原始3D视频信号进行信号分离，将原始3D图像信号转换成左图像信号、右图像信号及与左图像信号和右图像对应的L/R同步信号；
    色彩转换及图像提升单元，用于将L图像和R图像的色彩空间图像转换成YUV色彩空间，对YUV色彩空间的L图像和R图像采用亮度分区缩放和分级插值来进行图像提升，转换成像素为M*N阵列的L和R图像；其中，M和N为正整数；
    分类插帧单元，用于在L/R同步信号控制下，对像素为M*N的L图像和R图像进行分类，依次将L图像缓存在一起形成L图像集，将R图像缓存在一起形成R图像集；分别对L图像集和R图像集进行插帧，形成帧频为f1的L图像集和帧频为f1的R图像集；将L图像集和R图像集中的图像依次输出、以及输出对应R标识信号和L标识信号；其中，f1为正整数；
    RGB图像转换单元，用于对L图像和R图像进行RGB图像转换，将YUV色彩空间的L图像和R图像对应转换成RGB色彩空间的L图像和R图像；
    第一编织输出单元，用于对RGB色彩空间的L图像和R图像中的像素按规律排列，对像素中的R、G、B分量按规则编织成LVDS信号格式，并在L图像的最早传输的四个像素中插入L标识，在R 图像的最早传输的四个像素中插入R标识，以帧频为2f1的速度交替输出。
  11. 根据权利要求10所述的裸眼3D显示系统，其特征在于，所述色彩转换及图像提升单元包括：
    色彩转换单元，用于将L图像和R图像的色彩空间图像转换成YUV色彩空间，将L图像和R图像的像素转化成YUV4:2:2的像素；
    亮度分类单元，用于对YUV色彩空间的图像信号中亮度Y进行分析归类，按照亮度值高低分为低亮度区域、中亮度区域及高亮度区域；所述低亮度区域的亮度小于第一临界值，所述中亮度区域的亮度在第一临界值和第二临界值之间，所述高亮度区域的亮度大于第二临界值；所述第一临界值小于第二临界值；
    分区缩放单元，用于采用亮度分区缩放来提升对比度，对低亮区域的图像进行亮度压缩，对中亮区域的图像进行亮度增强，高亮度区域的亮度保持不变；
    分级插值单元，用于采用分级插值来提升分辨率，对不同的亮度区域来采用不同的插值算法，将图像分辨率转换为固定分辨率，转换后L图像和R图像的像素为M*N阵列。
  12. 根据权利要求9所述的裸眼3D显示系统，其特征在于，所述第二3D视频处理模块包括：
    第二3D视频处理单元，用于对所述LVDS信号格式图像信号进行解编织及图像分离，分离出L图像序列和R图像序列；获取L图像序列和R图像序列中物体对象的深度信息及估算出每个像素点的 深度信息，形成连续平滑的深度图；根据L图像和对应的深度图估算出多个L视点图像，根据R图像和对应的深度图估算出多个R视点图像；对多个L视点图像和多个R视点图像进行交织处理后形成包含多个视点的合成图像；对所述合成图像进行格式编织得到对应的VBO信号格式图像信号。
  13. 根据权利要求12所述的裸眼3D显示系统，其特征在于，所述第二3D视频处理单元包括：
    解编织及图像分离单元，用于将LVDS信号格式图像信号转换成RGB图像信号、分离出L图像和R图像；
    深度图生成单元，用于获取L图像和R图像中相似的物体对象，根据L图像和R图像中相似物体对象的水平位移，计算出L图像和R图像中相似物体对象的视差；根据L图像和R图像物体对象的色彩、纹理及光影信息进行处理，估算出每个像素点的深度信息，形成连续平滑的深度图；
    多视点生成单元，用于采用内插视点或外延视点的方式，根据L图像和深度图估算出多个L视点图像，根据R图像和深度图估算出多个R视点图像；
    视点合成单元，用于根据裸眼3D显示器的物理像素排列情况，对多个L视点图像及多个R视点图像进行交织处理后合成包括多个视点信息的合成图像；
    第二编织输出单元，用于对包含多个视点信息的合成图像进行格式编织，将其编织成对应的VBO信号格式图像信号，以频率为f2的 速度输出，所述f2为正整数。
  14. 根据权利要求13所述的裸眼3D显示系统，其特征在于，所述第三3D视频处理模块包括：
    解编织单元，用于对所述VBO信号格式图像信号进行解编织，将VBO信号格式的图像信号转换成RGB信号格式的合成图像；
    插帧倍频单元，用于对合成图像进行插帧和倍频处理，将一帧图像进行重复后变为相同的T帧图像，从而得到频率为Tf2的合成图像，其中，T大于1；
    图像校准单元，用于根据裸眼3D显示器的电气特性及显示效果要求，对合成图像进行图像校准；
    编织输出及显示单元，用于将校准后的合成图像编织为对应的VBO信号格式图像信号，以频率为Tf2的速度传输给裸眼3D显示器，以实现裸眼3D显示。
PCT/CN2016/073084 2016-01-20 2016-02-01 一种裸眼3d显示方法及系统 Ceased WO2017124586A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
AU2016304612A AU2016304612B2 (en) 2016-01-20 2016-02-01 Naked-eye 3D display method and system thereof
US15/507,494 US10326974B2 (en) 2016-01-20 2016-02-01 Naked-eye 3D display method and system thereof

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610036403.0A CN105681777B (zh) 2016-01-20 2016-01-20 一种裸眼3d显示方法及系统
CN201610036403.0 2016-01-20

Publications (1)

Publication Number Publication Date
WO2017124586A1 true WO2017124586A1 (zh) 2017-07-27

Family

ID=56301974

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/073084 Ceased WO2017124586A1 (zh) 2016-01-20 2016-02-01 一种裸眼3d显示方法及系统

Country Status (4)

Country Link
US (1) US10326974B2 (zh)
CN (1) CN105681777B (zh)
AU (1) AU2016304612B2 (zh)
WO (1) WO2017124586A1 (zh)

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106469039B (zh) * 2016-09-28 2019-05-03 华为技术有限公司 图像显示方法及系统
CN106888373A (zh) * 2017-03-29 2017-06-23 京东方科技集团股份有限公司 一种三维图像的处理装置及处理系统
CN107229337B (zh) * 2017-06-16 2020-08-18 上海闻泰电子科技有限公司 Vr显示系统及方法
CN110113596B (zh) * 2018-02-01 2021-04-27 上海济丽信息技术有限公司 一种可切换光栅式裸眼3d显示系统及显示方法
CN108765485A (zh) * 2018-04-10 2018-11-06 深圳康得新智能显示科技有限公司 图像的展示、处理方法及装置
CN108881878B (zh) * 2018-05-28 2021-04-20 张家港康得新光电材料有限公司 一种裸眼3d显示设备和方法
CN110022452B (zh) * 2019-05-16 2021-04-30 深圳市芯动电子科技有限公司 一种适于全息显示装置的视频抽帧方法及系统
CN112188182A (zh) * 2019-07-02 2021-01-05 西安诺瓦星云科技股份有限公司 立体显示控制系统、立体显示系统及立体显示控制器
CN112584125B (zh) * 2019-09-30 2024-07-30 北京芯海视界三维科技有限公司 三维图像显示设备及其显示方法
WO2021108771A1 (en) 2019-11-28 2021-06-03 Braid Health Inc. Medical imaging processing system
CN111741303B (zh) * 2020-06-09 2022-03-25 Oppo广东移动通信有限公司 深度视频处理方法、装置、存储介质与电子设备
CN115442580B (zh) * 2022-08-17 2024-03-26 深圳市纳晶云实业有限公司 一种便携式智能设备裸眼3d图片效果处理方法
US20240098284A1 (en) * 2022-09-15 2024-03-21 Illinois Tool Works Inc. Systems and methods for digital image compression
TWI826033B (zh) * 2022-10-07 2023-12-11 宏碁股份有限公司 影像顯示方法與3d顯示系統
CN115941925A (zh) * 2023-03-14 2023-04-07 江西科骏实业有限公司 3d显示方法、装置、设备及存储介质
CN118337978A (zh) * 2024-04-24 2024-07-12 广东金鼎移动传媒有限公司 一种基于most光纤网络的裸眼3d成像ai转换方法
CN119031114B (zh) * 2024-09-19 2025-11-18 北京睿视界健康科技有限公司 一种裸眼3d的显示方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20050100895A (ko) * 2004-04-16 2005-10-20 한국전자통신연구원 영상시점 변환 기능을 이용한 입체영상 변환 표시장치 및그 방법과 그를 이용한 다시점 입체영상 디스플레이 시스템
CN102196294A (zh) * 2010-03-16 2011-09-21 康佳集团股份有限公司 立体电视的信号处理方法及立体电视
US20120098942A1 (en) * 2010-10-26 2012-04-26 Thomas John Meyer Frame Rate Conversion For Stereoscopic Video
CN102752616A (zh) * 2012-06-20 2012-10-24 四川长虹电器股份有限公司 双目立体视频转换多目立体视频的方法
CN102802004A (zh) * 2012-08-15 2012-11-28 上海易维视科技有限公司 裸眼3d模组
CN103260044A (zh) * 2013-05-10 2013-08-21 深圳创维－Rgb电子有限公司 3d超高清信号处理方法和装置

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6108005A (en) * 1996-08-30 2000-08-22 Space Corporation Method for producing a synthesized stereoscopic image
US7733421B1 (en) * 2005-12-05 2010-06-08 Marvell International Ltd. Vector interpolator
CN101702780A (zh) * 2009-09-07 2010-05-05 无锡景象数字技术有限公司 多视点立体内容的参数化合成方法
JP5002702B2 (ja) * 2010-12-03 2012-08-15 株式会社東芝 視差画像生成装置、立体映像表示装置及び視差画像生成方法
JP2012138655A (ja) * 2010-12-24 2012-07-19 Sony Corp 画像処理装置及び画像処理方法
CN102215405B (zh) * 2011-06-01 2013-10-23 深圳创维－Rgb电子有限公司 3d视频信号压缩编解码方法、装置及系统
JP2013005259A (ja) * 2011-06-17 2013-01-07 Sony Corp 画像処理装置、および画像処理方法、並びにプログラム
JP2013038602A (ja) * 2011-08-08 2013-02-21 Sony Corp 画像処理装置、および画像処理方法、並びにプログラム
US9786253B2 (en) * 2012-01-25 2017-10-10 Lumenco, Llc Conversion of a digital stereo image into multiple views with parallax for 3D viewing without glasses
US9756316B2 (en) * 2013-11-04 2017-09-05 Massachusetts Institute Of Technology Joint view expansion and filtering for automultiscopic 3D displays
KR101980275B1 (ko) * 2014-04-14 2019-05-20 삼성전자주식회사 다시점 영상 디스플레이 장치 및 디스플레이 방법
CN104702935A (zh) * 2015-03-31 2015-06-10 王子强 一种基于视差补偿的裸眼3d显示方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20050100895A (ko) * 2004-04-16 2005-10-20 한국전자통신연구원 영상시점 변환 기능을 이용한 입체영상 변환 표시장치 및그 방법과 그를 이용한 다시점 입체영상 디스플레이 시스템
CN102196294A (zh) * 2010-03-16 2011-09-21 康佳集团股份有限公司 立体电视的信号处理方法及立体电视
US20120098942A1 (en) * 2010-10-26 2012-04-26 Thomas John Meyer Frame Rate Conversion For Stereoscopic Video
CN102752616A (zh) * 2012-06-20 2012-10-24 四川长虹电器股份有限公司 双目立体视频转换多目立体视频的方法
CN102802004A (zh) * 2012-08-15 2012-11-28 上海易维视科技有限公司 裸眼3d模组
CN103260044A (zh) * 2013-05-10 2013-08-21 深圳创维－Rgb电子有限公司 3d超高清信号处理方法和装置

Also Published As

Publication number Publication date
AU2016304612A1 (en) 2017-08-03
US10326974B2 (en) 2019-06-18
CN105681777A (zh) 2016-06-15
AU2016304612B2 (en) 2018-05-24
CN105681777B (zh) 2017-09-05
US20180109772A1 (en) 2018-04-19

Similar Documents

Publication Publication Date Title
CN105681777B (zh) 一种裸眼3d显示方法及系统
US8384766B2 (en) Apparatus for processing a stereoscopic image stream
US9924153B2 (en) Parallel scaling engine for multi-view 3DTV display and method thereof
CN103119947B (zh) 用于校正立体图像中的误差的方法和设备
CN102055998B (zh) 图像处理装置和图像处理方法
TWI428007B (zh) 視訊處理裝置及方法
US20040252756A1 (en) Video signal frame rate modifier and method for 3D video applications
CN101841685B (zh) 图像信号处理设备和方法、三维图像显示设备、传送/显示系统
CN1741621A (zh) 产生立体图像信号的方法和缩放该信号的方法
CN104604222B (zh) 用于产生、传送和接收立体图像的方法及有关设备
JP2012070032A5 (zh)
CN1169641A (zh) 立体观看装置和方法
CN102291587B (zh) 全高清3d视频处理方法
CN101015220A (zh) 三维图像再现设备
US20120140029A1 (en) Image Processing Device, Image Processing Method, and Program
CN102752620A (zh) 3d视频的电视播放方法
CN102257829B (zh) 立体影像显示装置及运动矢量导出方法
US20130050183A1 (en) System and Method of Rendering Stereoscopic Images
US20120120190A1 (en) Display device for use in a frame sequential 3d display system and related 3d display system
US20120120191A1 (en) Image processor for use in a frame sequential 3d display system and related 3d display system
WO2016004715A1 (zh) 显示处理系统、方法及电子设备
JP2014072809A (ja) 画像生成装置、画像生成方法、画像生成装置用プログラム
CN103260044B (zh) 3d超高清信号处理方法和装置
US8896663B2 (en) Transmitting device, receiving device, communication system and program
Kim et al. Object-based stereoscopic conversion of MPEG-4 encoded data

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 15507494

Country of ref document: US

ENP Entry into the national phase

Ref document number: 2016304612

Country of ref document: AU

Date of ref document: 20160201

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16885859

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16885859

Country of ref document: EP

Kind code of ref document: A1