SMaRT: the Smart Meeting Room Task at ISL

Abstract

As computational and communications systems become increasingly smaller, faster, more powerful, and more integrated, the goal of interactive, integrated meeting support rooms is slowly becoming reality. It is already possible, for instance, to rapidly locate task-related information during a meeting, filter it, and share it with remote users. Unfortunately, the technologies that provide such capabilities are as obstructive as they are useful -they force humans to focus on the tool rather than the task. Thus the veneer of utility often hides the true costs of use, which are longer, less focused human interactions. To address this issue, we present our current research efforts towards SMaRT: the Smart Meeting Room Task. The goal of SMaRT is to provide meeting support services that do not require explicit human-computer interaction. Instead, by monitoring the activities in the meeting room using both video and audio analysis, the room will be able to react appropriately to users' needs and allow the users to focus on their own goals.

A Survey on Smart Meeting Rooms and Open Issues

“…From the analyses of several projects we are able to categorize them in two sections: the projects that focus on the Operational Layer which only cover Pervasive and Ubiquitous Computing, HCI, Sensors and Networks areas, who approach the problem based on an Information System approach [10] [14]; and the ones that are also targeting the Intelligent Layer which propose developments in semantic processing and thus falling down on the Artifici al Intelligence (AI) area [6,9,11,13,[15][16][17][18][19][20]26].…”

3D tracking and dynamic analysis of human head movements and attentional targets

“…There has been a great interest in intelligent spaces that monitor the occupants and use head pose to measure their activities and visual focus of attention [2,[5][6][7][8][9][10][11][12][13]. Our proposed system differs from these works in many ways, most significantly in that we simultaneously estimate the head pose of four to five persons and utilize head pose estimation and tracking algorithms that provide a six degree-of-freedom interpretation of head position and orientation.…”

Automatic analysis of multiparty meetings

“…The first substantial effort to build a research infrastructure for multi-party meetings came with the work of Carnegie Mellon University (Waibel et al 2003) and the International Computer Science Institute (ICSI) , both of whom focused on speech recognition in meetings, primarily recorded using headset microphones. ICSI collected, transcribed and released a 75-hour corpus of multiparty meetings .…”

“…DRAFT Although, Uchihashi et al (1999) developed a novel approach to browsing such meetings based on video keyframes presented as a manga-style comic book, useful semantic search across meeting recordings relies on a transcription of what was said (as recognised by Kazman). Research teams at Carnegie Mellon University (Waibel et al 2003) and the International Computer Science Institute were the first to focus on the problem of speech recognition in meetings, as a key Meeting recognition and analysis presents a substantial set of interdisciplinary research problems, reflected in the above-mentioned projects. Meetings take place in a natural setting and their analysis must take account of several underlying characteristics:…”