Multi-view Imaging System Based On Kinect

Multi-view image technology has been in an increasing number of application ofcomputer vision. Traditional multi-view image is based on multi-camera array whichis large and heavy to operate. As the depth camera’ application is spreading, anmethod combining an depth camera with a color camera has attracted more and moreattention. The paper introduce the stereo imaging and display technology, as well asthe development of abroad at the first part. Then we propose a multi-view systembased on Microsoft Kinect. The depth map generated by Kinect is filtered by anedge-guide propagate trilateral filter, and inpainted depth map combined withcorresponding color map are used to generate virtual views via depth image basedrending technology. The raw virtual views exist large missing areas in background sothat we proposed a depth-guide priority method for searching block for inpainting firstand the use the matrix completion and matrix recovery model to solve the finalproblem. The main contributions are listed below:The paper proposed a multi-view image system based Microsoft Kinect RGB-Dcamara which is different from traditional stereo system. Before the depth map is usedto generate virtual views, it is preprocessed by an edge-guide propagate trilateral filterwhich takes account of color、depth、spatial distance information. The background andforeground is perfectly segmented and missing fields are inpainted.An block priority calculation method for virtual views inpainting is proposed.The method take advantage of depth information at different plane and texture、structure of color map. As the priority block is searched out along the border ofmissing area, we consider the problem as matrix completion and matrix recovery sothat an Enhanced Lagrangian operator is made use to inpaint virtual color map. Thefinal inpainted results are presented in two kinds of3D displays, one is two-viewstereoscopic display and another is auto-stereoscopic display. Strong3D experiencecan be felt in display.