Emotion detection in real-time on an Android Smartphone
CovenanteeIllinois Institute of Technology
Document typeMaster thesis
Rights accessOpen Access
During the last decade the proliferation and capabilities of smartphones have exploded, reaching a situation in which, nowadays, almost everybody carries one of these devices with enormous computational power, cameras and several sensors at any time. The omnipresence, the capabilities and the tools available to facilitate developing applications are the perfect combination for proposing solutions to numerous problems, some existing and some created. In this research, an approach of the widely researched topic of automatically detecting human emotions is brought to the Android platform by means of an application. A system is developed using some existing tools in order to detect emotions with few computations so that it may run in real time on smartphones with less computational power. Given the limited time available for developing this project, the scope of this work is to set the foundation so that the application can be improved in the future by other researchers; therefore, some limitations are set, and some stages are not fully optimized. Finally, future improvements are proposed to facilitate the continuation of this project.
SubjectsAndroid (Electronic resource), Computer vision, Machine learning, Android (Recurs electrònic), Visió per ordinador, Aprenentatge automàtic
DegreeMÀSTER UNIVERSITARI EN ENGINYERIA DE TELECOMUNICACIÓ (Pla 2013)