| With the rapid development of science and technology,the research of augmented reality technology has become more and more mature,and many researchers have begun to apply augmented reality technology to daily life.In normal life,when buying furniture,people can only buy furniture in the furniture store,and cannot see the scene of the furniture actually placed in their homes.Once they return to the home and find that the furniture is not suitable,they can only go back and proceed.Replacement greatly affects work efficiency.However,virtual placement of furniture in a furniture store through augmented reality technology,and the use of augmented reality equipment to view the display effect,can reduce the number of commodity replacements and improve work efficiency.Aiming at the realization of virtual furniture placement,this article has done the following research contents:Firstly,most augmented reality applications currently use VSLAM(Visual Simultaneous Localization and Mapping)to realize the registration of virtual objects.For VSLAM,the device moves too fast and the feature points between adjacent frames are too few,which leads to poor matching effects and affects the registration accuracy of virtual objects.Binocular vision SLAM and IMU(Inertial measurement unit)fusion method for tracking registration.Through the IMU pre-integration module,when feature point matching is performed at the front end of the augmented reality registration algorithm,the search space for matching feature points is reduced,the number of iterations is reduced,the matching process is accelerated,and the Algorithm efficiency: When the camera pose is determined at the back end of the algorithm,IMU data is added to jointly solve the camera pose,and the error term is added as a weighting factor to the solving equation to improve the accuracy of the solved camera position.Secondly,in view of the lack of shadow information and lack of realism in the virtual objects registered by the traditional virtual registration method,a method of constructing shadow bodies based on shadow regions to achieve the lighting consistency of virtual objects is proposed.Introduce and improve the super pixel segmentation algorithm.By adding pixel neighborhood information to calculate the color distance,the number of super pixel sets is reduced,and then a threshold is established to divide the image into shadow and non-shadow areas,and the shadow area is used to calculate the lighting parameters and construct Shadow volume,the constructed shadow volume can perform light and shadow rendering on the inserted virtual object to achieve the lighting consistency of the registered virtual object.Finally,based on the improved algorithm previously studied,using the publishable multi-platform application feature of the Visual Studio 2019 platform,a mobile-based augmented reality virtual furniture placement system is developed.In order to ensure a smooth experience of augmented reality interaction under the condition of limited mobile terminal resources,the rendering process is optimized.In 3Dmax,the vertices and triangles of the virtual furniture model are deleted,so that the number of triangles for constructing the shadow volume is drastically reduced.The number of triangles is greatly reduced,which greatly relieves the computing pressure of mobile device hardware and saves a lot of rendering time.Finally,the actual operating results of the system show that compared to some augmented reality systems of the same category,the system has higher registration accuracy and the registered virtual furniture has shadow information. |
PDF Full Text Request