Serious Games, Interaction and Simulation. 6th International Conference, SGAMES 2016, Porto, Portugal, June 16-17, 2016, Revised Selected Papers

Research Article

VirtualSign Game Evaluation

Download
217 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-51055-2_15,
        author={Paula Escudeiro and Nuno Escudeiro and Marcelo Norberto and Jorge Lopes},
        title={VirtualSign Game Evaluation},
        proceedings={Serious Games, Interaction and Simulation. 6th International Conference, SGAMES 2016, Porto, Portugal, June 16-17, 2016, Revised Selected Papers},
        proceedings_a={SGAMES},
        year={2017},
        month={1},
        keywords={VirtualSign Serious game Results Sign language Translator},
        doi={10.1007/978-3-319-51055-2_15}
    }
    
  • Paula Escudeiro
    Nuno Escudeiro
    Marcelo Norberto
    Jorge Lopes
    Year: 2017
    VirtualSign Game Evaluation
    SGAMES
    Springer
    DOI: 10.1007/978-3-319-51055-2_15
Paula Escudeiro1,*, Nuno Escudeiro1,*, Marcelo Norberto1,*, Jorge Lopes1,*
  • 1: Instituto Superior de Engenharia do Porto
*Contact email: pmo@isep.ipp.pt, nfe@isep.ipp.pt, matno@isep.ipp.pt, jolbl@isep.ipp.pt

Abstract

This paper presents the evaluation methods and techniques applied to the serious game developed within the VirtualSign project as well as the results achieved through those methods. VirtualSign is a Portuguese sign language bi-directional translator with three main components. The components are a gesture to text translator, a text to gesture translator and a serious game. The serious game aims to make the process of learning sign language easier and enjoyable using the VirtualSign bi-directional translator. The method used to evaluate the VirtualSign was the Quantitative Evaluation Framework (QEF). The translator undertakes a rigid validation process by both sign language experts and the deaf community. The evaluation process was also supported by questionnaires. The precision of the translator has a direct impact on the game performance. To evaluate the precision of the automatic translator we have used QEF and 10-fold cross validation to estimate the accuracy of the translator. The experimental results show a precision above 90%. In the identification of the Portuguese Sign Language terms As for the game the quality rate is 88% based on the QEF evaluation.