Real-Time Person Identification System for Intelligent Digital TV

Abstract

Intelligent digital TV (iDTV) is an enhanced digital TV that can automatically provide user-personalized services for each audience. For the user-personalized services, the iDTV should recognize audiences in real-time. Thus, in this paper, we define a novel structure of the iDTV and propose a real-time person identification system embedded in the iDTV. The proposed system consists of three phases: preprocessing for reducing computational costs of the proposed system, face detection using a statistical approach with Haar-like features, and face recognition using Support Vector Machines (SVMs). Experimental results show that the proposed system achieves efficient performance with high recognition accuracy of above 90% at the speed of 15~20 fps, which is suitable for applying to the iDTV.

Rough Set Theory Based User Aware TV Program and Settings Recommender

“…For the user personalized services, the iDTV should recognize audiences in real-time. In this paper, we define a novel structure of the iDTV and propose a real-time person identification system in the iDTV that analyzes captured images and recognizes audiences (Hwang, Ha, Kim, & Ko, 2007).…”

“…Consequently, additional technologies and applications have become available for consumers, and systems have become more intelligent by exploiting user information such as profiles, history, feedback, preferred program schedule, and the real-time program information offered by EPG and TV Anytime specifications. However, although these technologies have improved how well digital TVs respond to users, they must also take into account group of users that may want simultaneous access to the device, in addition to the technological achievements (Hwang, Ha, Kim, & Ko, 2007;Yu & Zhou, 2004;Yoon & Woo, 2008;Dewan:Sunil, 2009, North, Hargreaves, & Hargreaves, 2004.…”

Family Aware TV Program and Settings Recommender

“…For the user personalized services, the iDTV should recognize audiences in real-time. In this paper, we define a novel structure of the iDTV and propose a real-time person identification system in the iDTV that analyzes captured images and recognizes audiences [1].…”

“…However, although these technologies have improved how well digital TVs respond to users, they must also take into account group of users that may want simultaneous access to the device, in addition to the above technological achievements. [1,2,3,4,5] The goal of our research is to develop a recommendation engine for TV which facilitates the system to act autonomously in the way the family members desires in single and multiuser environment. We are the first to use the hierarchical structure of family effectively in recommendation engine.…”