Non-deforming three-dimensional image parallax quick adjusting method

Abstract

The invention relates to a non-deforming three-dimensional image parallax quick adjusting method which is characterized by comprising the following steps that S01. a depth value to be adjusted is determined, namely a depth changing value DA; S02. a parallax adjusting coefficient k = f(DA) is computed; S03. according to the smallest field depth changing error threshold value SDA = f(v and d) which can be perceived by a human eye, the value range (k1 and k2) of the parallax adjusting coefficient k is computed, wherein v is the watching distance between a person and a screen, and d is the distance between a three-dimensional image and the screen; and S04. an original image is zoomed according to a fixed drawing coefficient k, the value range of k is limited in (k1 and k2), and the fact that deforming cannot be perceived is guaranteed. Compared with a traditional method, the deforming degree is lowered by more than 50%, three-dimensional vision subjectivity effect is better, and method efficiency for high-definition video is improved by 48%.

Description

technical field [0001] The invention relates to the technical field of parallax adjustment, in particular to a method for quickly adjusting the parallax of a distortion-free stereoscopic image. Background technique [0002] At present, some research results have been obtained on the depth perception adjustment technology of 3D stereoscopic video at home and abroad. J. Konrad [2] It is first proposed to estimate the original depth through two views, on this basis, adjust the depth with proportional weight, and finally synthesize a new view. There is some image distortion in this method. D. Kim et al. [3] Add the visual fatigue prediction method before adjusting the depth, and use the method based on view synthesis to adjust the parallax when the visual fatigue occurs in the video content. Chang Yuan et al. [4] A method for adjusting 3D stereoscopic content according to the viewer's comfort and preference is proposed. First, the disparity is estimated by a block matching...

AI Technical Summary

Problems solved by technology

[0005] It can be concluded that there are problems in the parallax adjustment technology based on translation and stretching. The problems include: the traditional translation-based method has a perceivable deformation of the graphics after the depth perception is adjusted.

At the same time, when the image is panned, there will be holes and information loss at the edge of the display

The stretching method will also cause the deformation of the original graphics, and the information on the left and right edges of the image will disappear after the depth adjustment.

Images