Naked eye 3D implementation method

Abstract

The invention provides a naked eye 3D implementation method. The method comprises the steps of obtaining a 2D image, and splitting the 2D image into a background role and a foreground role; carrying out hierarchical processing on the image data obtained after splitting processing, and carrying out basic 3D projection on the image data obtained after hierarchical processing within a set size in a naked eye 3D camera angle coverage range; performing detail repair processing on the image data obtained after basic 3D projection by using OPEN CL 3D; and performing image calculation of 16 channels of the naked eye 3D image on the image data subjected to detail repair processing by using an Open CL multi-rendering mechanism to obtain the naked eye 3D image. By applying the method provided by the embodiment of the invention, the limitation of the existing 3D process can be overcome, and the effective naked eye 3D image can be manufactured. Strength of the workstation can be utilized to the maximum extent, and feasibility of technicians in controlling and managing all things in the workstation is achieved under the condition that additional renderers and procedures are not used.

Description

technical field [0001] The invention relates to the technical field of 3D images, in particular to a method for realizing naked-eye 3D. Background technique [0002] 3D image is a stereoscopic image seen through parallax. Using the principle of human binocular stereo vision, the audience can obtain three-dimensional space images from the video medium, so that the audience has an immersive feeling. [0003] At present, 3D displays or TVs that require glasses require two image channels for the left eye and the right eye. This 3D technology can be completed with simple 2D or 3D calculation software, and there are many software for this 3D technology circulating in the market. [0004] However, the core of the naked-eye 3D method is to go beyond the eyes and display images from different angles according to the position of the audience, thereby creating the illusion of three-dimensional images. In order to achieve the goal of naked-eye 3D, the hardware must have 16 images from v...

AI Technical Summary

Problems solved by technology

[0006] The disadvantages of the first naked-eye 3D implementation in the above-mentioned prior art are: lack of three-dimensionality, multiple digital distortions on the left and right, and require a lot of modification work

Although it is a lightweight 2D calculation method, it also needs to perform 8 corrections (8 times each for L and R) after performing the 16-body calculation of the L8 channel and the R8 channel, so the workload will increase, and the production efficiency and cost performance will increase. not tall

Images