Material Fusion Technology Based On2D Stereoscopic Video Streams

In recent years, with the rapid development of3D technology,3D video will replace2D videoto bring a unique visual feast. Depth information of the increased3D video scene, makes visualsense of realism and a sense of reality to enhance the number of favorite so numerous film andtelevision companies have3D video as the next step in the development of goals. The keytechnologies of3D video including content creation, compression, encoding and decoding,transmission, and terminal display for3D video. People are constantly committed to the studyof3D video technology, makes3D video technology has been developed quite mature, but inthe development process of3D video there is a problem has gradually highlighted and concern,that is the problem of scarcity of3D content.3D content production has become one of the keyfactors restricting the development of3D video.In order not to affect the stereo display effect of3D video, and reduce3D video productioncycle and production costs, the existing3D content production method has the stereoscopicphotography, computer generated3D animation and2D to3D. But since the complicateimplementation of stereoscopic photography, computer-generated time-consuming, the effect of2D turn to the3D is poor, and many other factors, these methods failed to solve the problem ofthe shortage of3D video content well.This paper presents a new3D video content production method, which is2D+3D stereo contentproduction method. In the method, we generate a new stereo video based on the fusion of theexisting2D video and stereo material, and the fusion technology of2D image and3D image isthe key technology to realize the2D+3D content production. Combine-perspective fusiontechnology and sub-perspective fusion technology are the two fusion technique proposed in thispaper about2D images and3D images. On the basic research of the above two kinds of FusionTechnology, this paper aimed at the angle of fusion technique presents a stereo combinationmethod used GPU to carry on the optimization. At the same time, this paper based on theexperiment, combined with the image fusion algorithm of the weighted average, proposes anadaptive fusion algorithm. This method not only makes the2D images and3D images Mooreinterference effect and transparent colored halo effect.