Image-based Modeling And Rendering Method Based On Camera Self-calibration

The demand of the reality of 3-D complex scene becomes more and more urgent accompanyed with the development of Virtual Reality(VR),and IBR scheme has appeared in this situation. The scheme has become an important research area in Computer Graphics in recent years. It has been widely applied to computer ViSC,computer vision,multimedia,and etc.The research of the thesis belongs to the field of IBR,and aims at applying camera self-calibration to IBR. Through a few images and some manual assistance,we can get the camera intrinsic and extrinsic parameters,and reconstruct scene,then use Warping method to draw at new view point.The whole procedure can be divided into three part:computing camera intrinsic and extrinsic parameters by self-calibration,get depth information through matching image corresponding points and using Warping to render image.In the self-calibration scheme,the thesis emphasizes the accuracy of camera intrinsic and extrinsic parameters.We presents an accurate F method based on corresponding point adjustment.The method adjusts coresponding points according to the fixedness of projective transformed cross ratio,then calculates F matrix accurately through linear and non-linear methods. When computing intrinsic parameter,A matrix,we simplify the step,and stress on the two important parameters of A.The result will be getten through solving Kruppa equation based on SVD decomposition.In order to compute extrinsic parameters,we use linear method to get initial R and t,then apply non-linear method to accurate them.In the scheme of getting depth information and reconstructing scene,the thesis presents two different methods:directly getting scene model through geometric information and getting depth information through matching corresponding points.Because the former is short of general purpose,we mainly discuss the latter. Based on previous research,the thesis presents a mothod which can pick up depth information through matching image corresponding points,and has getten better result.In the scheme of image Warping.the thesis discusses filling hole,Warping based on self-calibration and filling hole by several images. The thesis presents some Warping results of reality scene images at last.