Digital smart kiosk project
暂无分享,去创建一个
The Digital Smart Kiosk is an “aware” information kiosk that detects and tracks prospective clients and conveys this awareness. A single-camera vision system detects and tracks people in the kiosk’s vicinityThe kiosk display contains an animated talking face, a live image from the camera, and a web browser that presents graphical and textual information. The animated talking face conveys awareness of clients and invites interaction by turning and watching prospective clients and by speaking to them. The Smart Kiosk is deployed at a local cafe where we are gathering usage data. Ke~vorcls Public kiosk, talking emotive avatar, machine vision, user interface design, information display. PNTRQDUCTION Computerized public ldosks are common sights at airports, shopping malls, and stadiums. If movies were any indication, we would interact with kiosks that see us and identify who we are, understand what we say, speak in clearly enunciated sentences, and possess friendly, “human” personalities. Our research group is periodically asked why kiosks do not have these capabilities yet. The answer is easy; although technologies exist for most of these capabilities, the technologies are not yet sufficiently robust and flexible enough for deployment in public kiosks. However, with advances in computer processing power, it is possible to create kiosks that, in real time, sense and respond to their environment. The Smart Kiosk project was initiated at Digital to explore how advances in computer technology could be applied to improve public kiosks. Today’s public kiosk has good multimedia display abilities but limited sensing capabilities. Information from a client is restricted to presses on a touchscreen or keypad and the use of magnetic cards. The presence of a client is detected by a button press or card swipe. The departure of a client can only be inferred from the lack of button presses or card swipes. This indirect method of locating the client has undesirable consequences. Petmission to make digitavhard copies of311 or part oftbis material for personal or classroom use is graded without fee provided that the copies -are not made or dii%uted for profit or commercial advantage, the copyright notice. tie title of tbe publication and its date appear, and notice is given that copy-i&t is by p ermission of the AChI, Inc. To copy otbekse. IO republish to post on scn’ers or to rexiistribute to Iii requires specitic permission ador fee. CM 9X Los Angeles CA USA copyright 3998 0-s9791-975-019814..$5.00 For example, a person may leave money at an automatic teller machine because the machine has no method of identifying a departure and hence can’t warn the client. The capability to locate a person can be used to improve the quality of an interaction. One thing we believe is important and lacking in current kiosks is the conveyance of “awareness”. When two people converse, they indicate their interest in the conversation by cues such as eye contact, gestures, and vocal utterances. We hypothesize that a kiosk that is aware of clients and can convey this awareness in a socially acceptable manner will be received better than a traditional kiosk. This document contains a description of the design, implementation, and public deployment of a prototype kiosk. The document is divided into three sections. First, a discussion of the project design criteria and how they were met. Second, a detailed description of the prototype kiosk and the software used to locate people, display awareness, and communicate. Third, a discussion of the deployment of the kiosk in a commercial setting for testing. The intent of this document is not to describe results obtained from the kiosk deployment. PROJECT DESIGN CRITERIA The goal of the Smart Kiosk project is to create public kiosks that sense their environment and engage clients in a personable interaction. To accurately judge the success or failure of our efforts, we determined that our kiosks should be deployed and tested outside of the laboratory in public locations. These requirements defined the three project design criteria: client awareness, information presentation to the client, and deployment.
[1] Keith Waters,et al. Computer Facial Animation, Second Edition , 1996 .
[2] Thoms M. Levergood,et al. DEC face: an automatic lip-synchronization algorithm for synthetic faces , 1993 .
[3] S. Kiesler,et al. A prisoner's dilemma experiment on cooperation with people and human-like computers. , 1996, Journal of personality and social psychology.
[4] Demetri Terzopoulos,et al. Realistic modeling for facial animation , 1995, SIGGRAPH.