Inferring search intents from remote control movement patterns: a new content search method for smart TV

Abstract

Smart TV has been an increasingly important multimedia device in recent years due to its powerful functionalities that facilitate consuming multimedia contents. In smart TV environment where an interface for controlling multimedia contents is very limited, how to minimize user efforts while interacting with multimedia contents (e.g., searching for a specific video or browsing through images) is the most critical issue. To address this issue, various approaches using voice communication, gesture control, and gaze tracking systems have been proposed; however, their performance is still not satisfactory. In this paper, a new approach for providing multimedia contents to users with minimized human efforts in smart TV environment is proposed. The proposed approach is based on the recent smart TV equipped with a standard mouse as a remote controller and exploits a novel algorithm to infer user's search intent from mouse movement patterns, thereby improving the speed and accuracy of the contents search in smart TV environment. The experimental results show the strength and potentials of the proposed approach and further research directions.