The Design Of Multiple View Video Fusion And Representation System

With the development of video sensor network technology, the problem of expressing multiview video information appears recently. How to transform and fuse multi-viewing angle video information to improve the effectiveness of the expression of interesting targets. A software system of multi-viewing video fusion and representation is designed, which can implement three kinds of video information expressions, i.e., three-dimensional (3D) sketch reconstruction, collaborative multiview targets tracking, and video collaging. The system effectively and complementary represents the multiview video information. The work of this paper can be summarized as follows.This paper analyzes functional and performance requirements of the system according to the characteristics of 3D sketch reconstruction, collaborative multiview targets tracking, and video collage. Based on these research, the architecture and functional modules of the system are determined and designed in detail. The system is finally implemented with C/C++ language with strong versatility and scalability. Our system is fairly universal and extensible.This paper tests the function and performance of the main modules of the system. The experimental results show that the implemented system satisfies the requirements of multiview video representation. However, the real time of some algorithms remains to improve. The integration of the different representations also needs to be improved.In this paper, the representation of multiview video information is achieved by implementing 3D sketch reconstruction, collaborative multiview target tracking, and video collage. With the system, the scene and targets of interests can be observed via virtual 3D sketch, and the interesting targets can be locked with collaborative tracking. In addition, the scene information can be shown through other different expression ways of multiview video.