Analysis On Quality Of Experience In Haptic Human-computer Interaction

Due to the development of multimedia technology such as audio and video,users have higher requirements for immersion and realism of human-computer interaction technology.Existing studies show that the applications of haptic perception in human-computer interaction can enrich the interaction details and effectively improve the immersion and realism of interaction.The researchers take the Quality of Experience(QoE)as an index to evaluate the user’s perceived experience.QoE is a user-centered quality evaluation indicator,which can reflect the real experience of users and contribute to the improvement of humancomputer technologies.Therefore,it is necessary to design an effective QoE evaluation method for typical application scenarios.Based on the typical application scenarios of human-computer interaction,this thesis builds a haptic-visual interaction framework and a haptic human-computer interaction QoE model.The contributions of this thesis are as follows:(1)To explore the influence factors of QoE in human-computer interaction,and analyze its key role in the haptic control scene,this thesis compares the similarities and differences between the influence factors of QoE in traditional human-computer interaction and the haptic balance ball VR game.According to the existing haptic interaction researches,haptic control is taken as a typical application scenario.Based on this application scenario,combined with the current ubiquitous definition of QoE,this thesis discusses the influencing factors of QoE and selects the key factors to provide a theoretical basis for the follow-up research.(2)Based on the key factors of haptic manipulation QoE proposed in(1),a haptic-visual interaction framework and a haptic interaction database are established.At present,there is no large-scale public database for the research of haptic interaction QoE.Based on the haptic manipulation scene,this thesis establishes a haptic-visual interaction framework,designs subjective evaluation experiments with the overall user satisfaction as the QoE indictor,and then establishes a haptic interaction subjective evaluation database.The database contains quantitative indicators of QoE influencing factors related to the system,task,and user,and provides the overall satisfaction score of users,which provides a research basis for further exploring the mapping relationship between various influencing factors and interactive QoE.(3)Based on the haptic interaction database established in(2),the interactive QoE evaluation model in haptic manipulation scene is constructed by machine learning algorithm.This thesis explores the correlation between the single key factor of QoE and the overall satisfaction of users on the proposed haptic interactive database.In the haptic manipulation scenario,the mapping between multiple key factors of QoE and the overall satisfaction of users is constructed,and a variety of machine learning algorithms are used to realize the QoE evaluation model of haptic interaction,and the performance of the model is verified and compared.Besides,based on the haptic interaction QoE evaluation model,the influence of a single key element of QoE on the performance of the model is analyzed.Based on the typical application scenarios of haptic operation,this thesis selects the key factors of QoE and constructs a haptic visual interaction framework.Throughs the subjective evaluation experiments,this thesis establishes a haptic interaction database and proposes an accurate human-computer interaction QoE prediction model based on the database.The model is helpful to the research of adaptive interaction control scheme and QoE optimization,and improve the immersion and realism of human-computer interaction.