Yaohua Wang, Zhengtao Huang, Rongze Li, Xinyu Yin, Min Luo, Zheng Zhang, Xu Sun
{"title":"A Comparative Study of Speculative Retrieval for Multi-Modal Data Trails: Towards User-Friendly Human-Vehicle Interactions","authors":"Yaohua Wang, Zhengtao Huang, Rongze Li, Xinyu Yin, Min Luo, Zheng Zhang, Xu Sun","doi":"10.1145/3404555.3404617","DOIUrl":null,"url":null,"abstract":"In the era of growing developments in Autonomous Vehicles, the importance of Human-Vehicle Interaction has become apparent. However, the requirements of retrieving in-vehicle drivers' multi-modal data trails, by utilizing embedded sensors, have been consid- ered user unfriendly and impractical. Hence, speculative designs, for in-vehicle multi-modal data retrieval, has been demanded for future personalized and intelligent Human-Vehicle Interaction. In this paper, we explore the feasibility to utilize facial recog- nition techniques to build in-vehicle multi-modal data retrieval. We first perform a comprehensive user study to collect relevant data and extra trails through sensors, cameras and questionnaire. Then, we build the whole pipeline through Convolution Neural Net- works to predict multi-model values of three particular categories of data, which are Heart Rate, Skin Conductance and Vehicle Speed, by solely taking facial expressions as input. We further evaluate and validate its effectiveness within the data set, which suggest the promising future of Speculative Designs for Multi-modal Data Retrieval through this approach.","PeriodicalId":220526,"journal":{"name":"Proceedings of the 2020 6th International Conference on Computing and Artificial Intelligence","volume":"61 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 2020 6th International Conference on Computing and Artificial Intelligence","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3404555.3404617","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
In the era of growing developments in Autonomous Vehicles, the importance of Human-Vehicle Interaction has become apparent. However, the requirements of retrieving in-vehicle drivers' multi-modal data trails, by utilizing embedded sensors, have been consid- ered user unfriendly and impractical. Hence, speculative designs, for in-vehicle multi-modal data retrieval, has been demanded for future personalized and intelligent Human-Vehicle Interaction. In this paper, we explore the feasibility to utilize facial recog- nition techniques to build in-vehicle multi-modal data retrieval. We first perform a comprehensive user study to collect relevant data and extra trails through sensors, cameras and questionnaire. Then, we build the whole pipeline through Convolution Neural Net- works to predict multi-model values of three particular categories of data, which are Heart Rate, Skin Conductance and Vehicle Speed, by solely taking facial expressions as input. We further evaluate and validate its effectiveness within the data set, which suggest the promising future of Speculative Designs for Multi-modal Data Retrieval through this approach.