TY - GEN
T1 - Virtual space teleconferencing
T2 - 2nd Asian Conference on Computer Vision, ACCV 1995
AU - Kishino, Fumio
AU - Ebihara, Kazuyuki
AU - Ohya, Jun
PY - 1996
Y1 - 1996
N2 - This paper describes a virtual space teleconferencing that proposed as an human-oriented telecommunication system. By using this system, participants at different sites can engage in a conference with the sensation of sharing the same space, and cooperative work can be performed among remotely located participants. More specifically, our system connects three different sites via a 1.5 Mbps ISDN now in commercial use. The system has two large screens that can provide real-time reproduction of 3-D whole body human images. Participants at the three different sites are able to feel as if they are all at the same site. In this experimental system, to achieve real-time reproduction, tape marks, a data glove and magnetic sensors are used. However, these tools are not appropriate for natural human communications. It is a final goal to develop a system that does not require such tools. We also describe a new facial expression detection method that does not need the tape marks, and real-time detection is realized by exploiting frequency domain data from the face image acquired by the CCD camera. Experiments show very promising results for accurate expression detection and for implementation of real-time hardware of the using propose method.
AB - This paper describes a virtual space teleconferencing that proposed as an human-oriented telecommunication system. By using this system, participants at different sites can engage in a conference with the sensation of sharing the same space, and cooperative work can be performed among remotely located participants. More specifically, our system connects three different sites via a 1.5 Mbps ISDN now in commercial use. The system has two large screens that can provide real-time reproduction of 3-D whole body human images. Participants at the three different sites are able to feel as if they are all at the same site. In this experimental system, to achieve real-time reproduction, tape marks, a data glove and magnetic sensors are used. However, these tools are not appropriate for natural human communications. It is a final goal to develop a system that does not require such tools. We also describe a new facial expression detection method that does not need the tape marks, and real-time detection is realized by exploiting frequency domain data from the face image acquired by the CCD camera. Experiments show very promising results for accurate expression detection and for implementation of real-time hardware of the using propose method.
UR - http://www.scopus.com/inward/record.url?scp=84949997231&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84949997231&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84949997231
SN - 9783540607939
VL - 1035
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 159
EP - 168
BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PB - Springer Verlag
Y2 - 5 December 1995 through 8 December 1995
ER -