Name
Abstract | ||
|---|---|---|
Smartphones are the most general ubiquitous computing devices available in common people's daily life. However, there are not so many applications for disabled, such as blind people. In this paper, based on internal acceleration sensors of mobile devices, we present a gesture recognition system, providing users with an original, direct form of human-computer interacting way. Unlike previous research on gesture recognition, our system, GRIB, can recognize user-defined gestures instead of pre-defined fixed ones. Thus, the system avoids laborious and time-consuming training set collection processes, and enhance the scalability of system. GRIB firstly utilizes displacement models to filter raw sensor sample, then matches with existed standard gestures defined by users themselves. We propose a quick and efficient algorithm to realize matching process. Our scheme can serialize each gesture using acceleration sensors and calculate the similarity between sample and defined gesture. To evaluate the performance of our system, we tested 25 gesture samples. The experiment result shows a high recognition rate about 96%. It shows our system can be wildly deployed in existing smart phones without additional hardware. The system can bring much convenience to users, especially for people with disabilities. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1109/CIT.2014.150 | CIT |
Keywords | DocType | ISSN |
mobile devices,internal acceleration sensors,human computer interaction,grib,gesture recognition interaction,gesture recognition, human-computer interaction, mobile devices, blind,ubiquitous computing devices,blind people,human-computer interacting way,smart phones,gesture recognition,gesture recognition system,handicapped aids,mobile computing,user-defined gestures,blind,human-computer interaction | Conference | 2474-9648 |
Citations | PageRank | References |
0 | 0.34 | 4 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yaodong Huang | 1 | 1 | 1.04 |
| Zhiyuan Xu | 2 | 1 | 1.04 |
| Ruijin Wang | 3 | 0 | 0.68 |
| Dajiang Chen | 4 | 78 | 8.24 |