Research On Visual Positioning Of Immersive Virtual Reality System In Large Space

With the deepening of computer technology research and the general application of Internet technology,virtual reality,or VR for short,has gradually emerged.People apply virtual reality to many traditional industries,which brings a new direction for the development of traditional industries.Immersive virtual reality is more popular because of its good experience.Manufacturers have launched a variety of helmets,glasses and other devices to meet the requirements of people's virtual experience,but there is still no mature system solution that can be fully applied to large-space immersive virtual reality systems.The key issue is the positioning of large spaces.In this paper,a low-cost and accurate indoor positioning system for large space is built by using computer vision and positioning signs.The main work includes:First of all,the overall implementation of the positioning system is introduced.The positioning system in this article is essentially a kind of indoor visual positioning system based on marks.It uses computer vision and location marks to perform positioning.The "two-dimensional code" location mark used in this paper can carry location information and be perfectly suitable for large space environment.Then the identification and detection process of the identification is described in detail.The camera image is collected under different light conditions,processed by different methods and compare the results of several methods.So we can use the camera for active visual positioning.Secondly,the camera pose estimation is described and a algorithm based on quaternion from least feature points is proposed in this paper,which can solve the camera coordinates by operation.The camera image is detected and the feature points are matched,and the equations are established by the constraint conditions of the feature points,so that the camera pose can be solved by constructing the eigenvalue problem.Compared the performance of this algorithm with several existing camera pose estimation algorithms,this algorithm has obvious advantages in the case of noise interference and more feature points available.At last,the specific design scheme of the system is described,the development framework and process of the system are introduced and realize each functional module inthe system and create a displacement interaction system between the virtual space and the realization world.Then Collect data and draw pictures at the same time to verify the practicability of the visual positioning system.