Search

CN-122027779-A - Data processing method and device for naked eye 3D display

CN122027779ACN 122027779 ACN122027779 ACN 122027779ACN-122027779-A

Abstract

The application discloses a data processing method and device for naked eye 3D display. The method comprises the steps of obtaining data to be processed, wherein the data to be processed comprises source data output by a received upper computer, performing virtual viewpoint generation processing based on depth information on the data to be processed to obtain multi-viewpoint image data, performing viewpoint optimization processing based on a human eye viewing model on the multi-viewpoint image data to obtain optimized viewpoint set data, and performing raster mapping processing on the optimized viewpoint set data to obtain naked eye 3D display data, wherein the naked eye 3D display data are data used for representing raster mapped image frames. And receiving the source data at the display end, performing raster mapping rendering on the source data, and then generating naked eye 3D display data, so as to realize complete link management of naked eye 3D data display.

Inventors

  • Song Chongguo

Assignees

  • 汇智天下(杭州)科技有限公司

Dates

Publication Date
20260512
Application Date
20260415

Claims (10)

  1. 1. The data processing method for naked eye 3D display is characterized by being applied to a naked eye 3D display system to realize 3D display data generation of received source data, and comprises the following steps: acquiring data to be processed, wherein the data to be processed comprises source data output by a received upper computer; Performing virtual viewpoint generation processing based on depth information on the data to be processed to obtain multi-viewpoint image data; Performing viewpoint optimization processing based on a human eye viewing model on the multi-viewpoint image data to obtain optimized viewpoint set data; And carrying out raster mapping processing on the optimized viewpoint set data to obtain naked eye 3D display data, wherein the naked eye 3D display data is data for representing raster mapped image frames.
  2. 2. The data processing method according to claim 1, wherein performing a virtual viewpoint generation process based on depth information on the data to be processed to obtain multi-viewpoint image data includes: performing identification processing based on source data on the data to be processed to obtain 2D image data and depth map data; Performing depth-based image deformation processing on the 2D image data and the depth map data to obtain image deformation data; And performing image optimization processing based on hole filling on the image deformation data to obtain the multi-view image data.
  3. 3. The data processing method according to claim 2, wherein performing hole filling-based image optimization processing on the image deformation data to obtain the multi-viewpoint image data includes: Performing hole detection processing on the image deformation data to obtain hole area data, wherein the hole area data is used for representing a hole area generated by image deformation; performing first cavity filling processing on the cavity area data to obtain first filling result data, wherein the first cavity filling processing is cavity area scanning filling; Performing second hole filling processing on the first filling result data to obtain second filling result data, wherein the second hole filling processing is hole area background filling; and performing third hole filling processing on the second filling result data to obtain the multi-view image data, wherein the third hole filling processing is hole area edge restoration.
  4. 4. The data processing method according to claim 1, wherein performing viewpoint optimization processing based on a human eye viewing model on the multi-viewpoint image data to obtain optimized viewpoint set data includes: Performing recognition processing based on viewing parameters on the data to be processed to obtain viewing parameter data, wherein the viewing parameter data are parameter data used for representing a user viewing distance and a user viewing angle; performing viewpoint weight distribution processing based on a human eye viewing model on the viewing parameter data to obtain viewpoint weight distribution data; Performing view optimization processing on the multi-view image data and the view weight distribution data to obtain optimized view set data, wherein the optimized view set data is set data for representing views after weight distribution.
  5. 5. The data processing method according to claim 4, wherein performing viewpoint optimization processing based on a human eye viewing model on the multi-viewpoint image data to obtain optimized viewpoint set data includes: the viewing parameter data are identified to obtain multi-user viewing parameter data, wherein the multi-user viewing parameter data are data used for representing viewing parameters corresponding to a plurality of users respectively; Performing viewing parameter equivalent processing on the multi-user viewing parameter data to obtain equivalent viewing parameter data; Performing viewpoint weight distribution processing based on a human eye viewing model on the equivalent viewing parameter data to obtain viewpoint weight distribution data; And performing view optimization processing on the multi-view image data and the view weight distribution data to obtain the optimized view set data.
  6. 6. The data processing method according to claim 1, wherein performing raster mapping processing on the optimized viewpoint set data to obtain naked eye 3D display data includes: Carrying out recognition processing based on grating parameters on the data to be processed to obtain grating parameter data, wherein the grating parameter data is data for representing physical parameters of a grating lens; performing geometric mapping model construction processing on the grating parameter data to obtain a grating mapping model, wherein the grating mapping model is used for representing the mapping relation between screen pixels and virtual viewpoints; Performing screen point location inverse mapping processing on the optimized viewpoint set data based on the grating mapping model to obtain screen point location pixel data, wherein the screen point location pixel data is pixel data for representing a virtual viewpoint corresponding to a screen point location; and obtaining the naked eye 3D display data according to the screen point position pixel data.
  7. 7. The data processing method according to claim 6, wherein performing the screen point location inverse mapping process based on the raster mapping model on the optimized view point set data, to obtain screen point location pixel data includes: Acquiring point location data of a display screen; performing pixel space discrete processing on the display screen point location data to obtain sub-pixel data; Performing viewpoint mapping processing based on the grating mapping model on each piece of sub-pixel data to obtain display viewpoint image data, wherein the display viewpoint image data is used for representing the image data of the display viewpoints of the sub-pixel data; and performing color extraction processing on the display viewpoint image data to obtain the screen point location pixel data.
  8. 8. The data processing device for naked eye 3D display is characterized by being applied to a naked eye 3D display system to realize 3D display data generation of received source data, and comprises the following components: The data acquisition module is used for acquiring data to be processed, wherein the data to be processed comprises source data output by a received upper computer; The parallax synthesis module is used for carrying out virtual viewpoint generation processing based on depth information on the data to be processed to obtain multi-viewpoint image data; the viewpoint optimizing module is used for performing viewpoint optimizing processing based on a human eye watching model on the multi-viewpoint image data to obtain optimized viewpoint set data; and the grating mapping module is used for carrying out grating mapping processing on the optimized viewpoint set data to obtain naked eye 3D display data, wherein the naked eye 3D display data is data for representing the image frames after grating mapping.
  9. 9. The naked eye 3D display system is characterized by comprising an upper computer, a grating mapping rendering module and a naked eye 3D display screen, wherein, The upper computer generates source data, codes the source data and transmits the source data to the grating mapping rendering module; The raster mapping rendering module is used for executing the data processing method for naked eye 3D display according to any one of claims 1-7; And the naked eye 3D display screen receives naked eye 3D display data output by the grating mapping rendering module so as to realize a naked eye 3D effect.
  10. 10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions for causing the computer to execute the data processing method for naked eye 3D display according to any one of claims 1 to 7.

Description

Data processing method and device for naked eye 3D display Technical Field The application relates to the field of 3D image display, in particular to a data processing method and device for naked eye 3D display. Background With the continuous development of technology, naked eye 3D technology has been developed as an attractive technology that allows a user to view 3D images or videos in a naked eye state without wearing any glasses or devices. The core principle is that left and right eye images are respectively projected to corresponding eyeballs through optical elements such as gratings, lens arrays or parallax barriers, and stereoscopic impression is formed by utilizing vision difference of human brains. The naked eye 3D display system in the prior art mainly depends on a high-configuration PC host (usually carrying RTX series independent display cards). The PC host takes on the task of full link from the construction of 3D scene, the generation of multi-view images, and the most power-consuming raster mapping. Finally, the PC transmits a "composite image that has completed the raster arrangement" to the display via the HDMI/DP cable. The result image after grating mapping is transmitted in the naked eye 3D display system in the prior art, and the applicant finds that the result image is formed by interweaving fragments with a very large number of sub-pixel levels, the requirement on the integrity of the image is very high, the image cannot be subjected to conventional video compression, when the image is subjected to encoding compression (such as network transmission or wireless screen projection) such as H.264/H.265, a compression algorithm can misjudge fine grating arrangement as noise points to smear, so that the 3D effect of a receiving end is completely invalid, and serious crosstalk, ghost and other problems occur. Therefore, the naked eye 3D display in the prior art has the problem of poor display effect. Disclosure of Invention The application mainly aims to provide a data processing method and device for naked eye 3D display, which are used for solving the technical problem that the naked eye 3D display in the prior art has poor display effect. In order to achieve the above object, a first aspect of the present application provides a data processing method for naked eye 3D display, which is applied to a naked eye 3D display system to implement 3D display data generation on received source data, and the method includes: acquiring data to be processed, wherein the data to be processed comprises source data output by a received upper computer; Performing virtual viewpoint generation processing based on depth information on the data to be processed to obtain multi-viewpoint image data; Performing viewpoint optimization processing based on a human eye viewing model on the multi-viewpoint image data to obtain optimized viewpoint set data; And carrying out raster mapping processing on the optimized viewpoint set data to obtain naked eye 3D display data, wherein the naked eye 3D display data is data for representing raster mapped image frames. Further, performing virtual viewpoint generation processing based on depth information on the data to be processed to obtain multi-viewpoint image data, including: performing identification processing based on source data on the data to be processed to obtain 2D image data and depth map data; Performing depth-based image deformation processing on the 2D image data and the depth map data to obtain image deformation data; And performing image optimization processing based on hole filling on the image deformation data to obtain the multi-view image data. Further, performing image optimization processing based on hole filling on the image deformation data to obtain multi-view image data includes: Performing hole detection processing on the image deformation data to obtain hole area data, wherein the hole area data is used for representing a hole area generated by image deformation; performing first cavity filling processing on the cavity area data to obtain first filling result data, wherein the first cavity filling processing is cavity area scanning filling; Performing second hole filling processing on the first filling result data to obtain second filling result data, wherein the second hole filling processing is hole area background filling; and performing third hole filling processing on the second filling result data to obtain the multi-view image data, wherein the third hole filling processing is hole area edge restoration. Further, performing viewpoint optimization processing based on a human eye viewing model on the multi-viewpoint image data to obtain optimized viewpoint set data includes: Performing recognition processing based on viewing parameters on the data to be processed to obtain viewing parameter data, wherein the viewing parameter data are parameter data used for representing a user viewing distance and a user viewing angle; performing viewpoint w