Research on Interactive Product Design Based on User Behavior Data in Virtual Reality Environment

Abstract

This paper designs an interactive product based on the virtual reality environment and related technologies and further optimizes the interactive product based on the user behavior data collected by the interactive product. Based on the ORB-SLAM algorithm, we construct a hand controller degree of freedom model for the interactive product to overcome the limitations of hand controller positioning. By constructing the ORB-SLAM3 jump perception model, the interactive product can be used more smoothly. The collected user interaction behavior data is downsized using PCA principal component analysis and interaction behavior characteristics of different users are classified using the LATM network. The average completion times of the camera proposed in this paper as a hand controller for tasks such as 3D pointing are 14.23, 12.29 and 13.68 seconds, respectively, which all perform well compared to other hand controllers. At the same time, the interactive products designed using the method of this paper have the least abnormalities detected by users during the use process, and the highest abnormal feeling rate is only 37.22%. There are significant differences in the behavioral characteristics of users during the use of interactive products, based on the distribution of feature data can be divided into three categories: virtual exploration, interactive experience, and traditional. Strategic directions for further optimizing interactive products can be provided through the behavioral analysis of users in different categories.