A smart TV system with body-gesture control, tag-based rating and context-aware recommendation

Smart TV, which enables viewers to conveniently access different multimedia contents and interactive services in a single platform, is currently becoming popular. The work presented here develops three new ways to enhance the performance of a smart TV. The first is to design a body control system that recognizes and interprets human gestures as machine commands to control the TV. The second is to create a new social tag-based method to recommend the most suitable multimedia contents for users. Finally, a context-aware platform is implemented to consider different environmental situations to make the best recommendations. Smart TV has quickly become popular in the market in recent years. In contrast to traditional TVs that only focus on media broadcasting, smart TV systems can deliver diverse multimedia contents from networked devices directly to the end-users, allowing them to access the contents through a user-friendly interface. Smart TVs also provides interactive Internet-based services, including media-on-demand, social networking, and on-line gaming. Smart TV continues to develop rapidly to offer increasingly more functions and services. In this work, we address three important issues closely related to smart TV, including the human–machine interface, content recommendations, and context awareness, and we develop the corresponding mechanisms to further enhance the performance of a smart TV system. Traditionally, people have used device-based control to operate different consumer electronics at home. Researchers then began to implement systems that utilize personal hand-held devices to work as controllers [1,2]. Recently, different methods of human– machine interactions (such as voice-based and gesture-based control) have been proposed to provide natural control on the equipment without any remote control devices [3,4]. Gesture control has many advantages over traditional control methods, such as its high accessibility (with no need of physical contact), high flexibility (adaptable to various applications), and low cost (fixed sensor hardware cost, in contrast to touchscreen cost which increases exponentially while the screen size increases). To move towards an even more natural way for users to interact with machines , in this work we design a body control mechanism through which the smart TV can respond to the human users' actions. On a smart TV system which can easily present different types of multimedia content to end-users, a large amount of contents can lead to information overload. Therefore, it is necessary to develop personalization techniques to recommend the most suitable content to users [5,6]. Many recommendation methods have been proposed, ranging from content-based …